Journal of the Toirey Botanical Society 133(1), 2006, pp. 83—118

Paleoecology of Late Paleozoic pteridosperms from tropical Euramerica^ William A. DiMichele^ ^ Department of Paleobiology, NMNH Smithsonian Institution, Wasliington, DC 20560 Tom L. Phillips Department of Plant Biology, University of Illinois, Urbana, IL 61801 Hermann W. Pfefferkorn Department of Earth and Environmental Science, University of Pennsylvania, Philadelphia, PA 19106

DiMlCHELE, W. A. (Department of Paleobiology, NMNH, Smithsonian Institution, Washington, DC 20560), T. L. PHILLIPS (Department of Plant Biology, University of Illinois, Urbana, IL 61801), AND H. W. PFEFFERKORN (Department of Earth and Environmental Science, University of Pennsylvania, Philadelphia, PA 19106). Paleo- ecology of Late Paleozoic pteridosperms from tropical Euramerica. J. Torrey Bot. Soc. 133: 83—118. 2006.— Late Paleozoic pteridosperms are a paraphyletic group of seed plants that were prominent elements of tropical ecosystems, primarily those of wetlands or the wetter portions of seasonally dry environments. Because the group is more a tradition-based, historical construct than a well circumscribed phylogenetic lineage, the wide variance in ecological roles and ecomorphological attributes should not be surprising. Pteridosperms can be the dominant canopy trees in local habitats, prominent understory trees, scrambling ground cover, thicket-formers, or liana-like plants and vines. Some species appear to have been weedy opportunists, although this ecological strategy seems to be a minor part of the wide spectrum of pteridosperm life habits. Most pteridosperms appear to have preferred wetter parts of the landscape, though not standing water, and relatively nutrient-rich settings (in comparison with groups such as tree ferns or lycopsids). Of the Paleozoic pteridosperms as traditionally circumscribed, only the peltasperms survived to become major elements in the Mesozoic. However, these plants may have been part of a derived seed-plant clade that also includes the corystosperms and cycads (see Hilton and Bateman, this volume), indicating that only the most derived of the Paleozoic pteridosperm lineages, those that appear to have evolved initially in extrabasinal settings, persisted into the Mesozoic. Key words: Carboniferous, Devonian, paleoecology. Paleozoic, Permian, pteridosperms, seed ferns.

The pteridosperms ("seed ferns") are an easy cortical regions. Many are vines or appear to group to caricature but difficult to characterize. have climbing plants at the root of their local Phylogenetically diverse, and denizens of a wide branch of the phylogenetic tree. And large to range of terra firma habitats, the pteridosperms super-sized seeds seem to have been prominent generally have been circumscribed by the pos- parts of the pteridosperm equation in some of session of "fern-like", compound fronds while the better known groups. Yet, there are clear ex- also being the bearers of seeds. All are woody, ceptions to all of these attributes that make the but most not prominently so, the wood often ap- group quite difficult to delineate ecologically. pearing to be an evolutionary afterthought rather Just as there is no typical pteridosperm mor- than the principal support tissue, except in a few phology (Galtier 1986), there likewise is no typ- clades. In addition, pteridosperms generally ical pteridosperm ecology, even though many of have monoaxial stems and sclerenchymatous the putative clades within the 'pteridosperms' are ecologically rather uniform and narrowly ' The authors wish to acknowledge the following distributed within the broader spectrum of eco- sources of support: The Evolution of Terrestrial Eco- logical resource space. systems Program of the Smithsonian Institution (WD) and the National Science Foundation, Earth Sciences Few Paleozoic pteridosperms persist into the Division, Grant EAR-0207848 (HWP). Mesozoic, with the exception of the peltasperms. - We thank Robert Gastaldo, Jason Hilton, and Dan Of several lineages, the meduUosans and lygi- Chaney for discussion about the ideas presented in this nopterids may be considered the "classic" Pa- manuscript and for sharing expertise and unpublished data. For detailed comments and suggestions on earlier leozoic seed fern groups, both best known from versions of the manuscript we are indebted to Richard the Pennsylvanian, in both adpression and pet- Bateman, Chris Cleal, Jean Galtier, and Hans Kerp. rifaction preservation. Other forms such as Cal- ' Author for correspondence: E-mail: dimichele@si. listophyton, the mariopterids, the peltasperms, edu Received for publication September 2, 2005, and in and most of the latest Devonian and Mississip- revised form October 18, 2005. pian seed-plant taxa are treated traditionally as

83 84 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133 pteridosperms largely because of their posses- among those groups traditionally identified as sion of some mixture of attributes, such as com- pteridosperms. Nonetheless, groups with shared plex, frond-like leaves, sclerotic cortex, or ra- morphological characteristics can be identified, dially symmetrical seeds. Unfortunately, the de- often with confidence, even though the subse- tailed relationships among these groups are rath- quent relationships among these "pteridosperm" er poorly understood at the level of genera and groups are not well understood in many cases. species, even though a general phylogenetic ar- Phylogenetic pattern is relevant to ecological chitecture for the group does exist (Rothwell and pattern because roles and environmental re- Serbet 1992, 1994; Hilton and Bateman, this sponses may be quite similar within clades and volume). can be quite different from one clade to another In this paper we review what is known or sus- Clade membership seems to specify the basic pected of the ecology of the Paleozoic pterido- ecological attributes of many families, genera sperms, focusing on the ecologies of the earliest and species. This is quite clear in Paleozoic eco- forms, the autecologies of the broad diversity of systems, especially those of the pre-Permian Pennsylvanian-aged groups, the place of pteri- wetlands (DiMichele and Phillips 1996a) where dosperms in the synecology of Pennsylvanian species richness is low and several Linnean tropical wetlands, and finally the ecological pat- class-level groups, each characterized by a dif- terns inferred for pteridosperms in the ensuing ferent basic body plan, dominate different eco- Permian seasonally dry tropical lowlands. The logical portions of landscapes. Such patterns general message is that pteridosperms were di- have been identified even in modern systems verse and ecologically heterogeneous, with a (Prinzing et al. 2001). For the late Paleozoic, the wide range of complex growth morphologies root of this extreme clade-by-environment eco- and roles in ecosystems of the time. logical partitioning lies in the Middle to Late A note on geological terminology used in this Devonian evolutionary radiation of basic plant paper: In order to date the events or fossil oc- body plans (DiMichele et al. 2001). The result- currences discussed herein, we will use the clas- ing pattern persisted until the Pennsylvanian, sical European Carboniferous stage names that when global environmental changes began to are most appropriate for terrestrial deposits. disrupt the established dominance hierarchy, These names, from youngest to oldest, are: Tour- leading to the rise of dominance by a variety of naisian, Visean, Namurian, Westphalian, and seed-plant groups throughout the world (Gastal- Stephanian. We will refer to the Tournaisian as do et al. 1996). "early" Mississippian and the Visean as "mid- In the case of the pteridosperms, certain pat- dle" Mississippian. The old, informal substages terns of clade-by-ecology congruence are evi- of the Namurian (A, B, C) also will be used, dent. Examples include the general liana-like where Namurian A (= Serpukhovian) is latest habit of the members of the Pennsylvanian- Mississippian (Early Carboniferous) and the oth- Permian Mariopteridaceae (Krings et al. 2003b) er two stages represent the earliest Pennsylva- and Pennsylvanian Lyginopteridaceae (Phillips nian (Late Carboniferous). The use of the more 1981, Speck 1994, Gastaldo et al. 2004), the numerous, formal substages of the Namurian scrambling to liana-like habit of the members of would be cumbersome and imply a precision the Mississippian Calamopityaceae (Galtier that is neither possible nor necessary in this re- 1992), the tree architectures of the woody lygi- view. We use Asselian, Sakmarian, Artinskian, nopteroids of the Mississippian (Galtier 1992), and Kungurian for the Early Permian, and Guad- and the gracile habit and opportunistic life his- alupian for the earliest stage of the Middle tory of the earliest seed plants (Rothwell and Permian. Scheckler 1988, Scheckler 1986b). Another in- stance is the vine/climber ancestry of the Med- Pteridosperm Phylogeny: Relevance to ullosales (Dunn et al. 2003b), which manifests Ecology. The pteridosperms are a paraphyletic itself in the persistent polystelic anatomy of the group as resolved by most phylogenetic analyses group. Such anatomy can be found both in spe- (Rothwell and Serbet 1994, Nixon et al. 1994, cies with weak-stemmed, leaning, thicket-form- Doyle 1996, Hilton and Bateman, this volume). ing habit and in those forms with upright, mono- However, because these phylogenetic analyses podial tree habit (Pfefferkorn et al. 1984, Wnuk focus mainly on a few representative members and Pfefferkorn 1984), some of which reached of a selected subset of seed-plant clades, they nearly 30 cm in diameter with massive second- provide only a general sense of the relationships ary xylem and adherent large leaf bases (Cotta 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 85

1832). At a somewhat broader level, the pelta- This was a precursor to the pollination-droplet sperms, although a highly diverse group of capture and pollen-tube sperm delivery systems plants, appear to have preferred stream-side, pe- of later gymnosperms, first recognized in the riodically dry to quite dry substrates. The Pennsylvanian (Rothwell 1981). Although hy- growth habits of peltasperms are not well un- drasperman prepollen capture mechanisms were derstood, but the best known had rather slender evolutionarily primitive, these wind delivery axes and small fronds or leaves (Kerp 1988, systems provided early seed plants with the po- Galtier and Broutin 1995, Barthel 2001). tential to escape the need for open-water sperm It is possible that delineation of greater phy- delivery. Thus, it gave them the potential to ex- logenetic structure in the pteridosperms will per- ploit environments in a new way, compared with mit the identification of more clade-level eco- their free-sporing heterosporous ancestors. Air- logical specificity than now is recognized, based flow patterns around early seed-plant ovules and on the patterns that can be identified to date. associated mechanisms of prepollen capture Greater phylogenetic structure also will permit were studied by Niklas (1981a, b), who dem- ecology to play a larger role in understanding onstrated that turbulent flow, produced by the evolutionary dynamics in this group. cupules, increased prepollen-grain impacts in the area of the micropyle. He further demonstrated The Earliest Pteridosperms. The earliest that younger, presumably more evolutionarily seed plant described as a "whole plant" is Elk- derived ovules had greater numbers of prepollen insia polymorpha, from rocks of mid-Late De- impacts. Ovules/seeds had open integuments, vonian age. This plant has been characterized as meaning that the integument consisted of lobes a seed fern in the broad sense (Serbet and Roth- that were unfused at, and some distance back well 1992) and may be considered broadly rep- from, the apex (Gillespie et al. 1981). Conse- resentative, in terms of its ecology, of many ear- quently, it seems reasonable to assume (though ly seed plants, part of an apparent radiation of an assumption it is) that the plants did not have this group in the Late Devonian. Some of these a mechanism for seed dormancy. early forms, such as Moresnetia zalesskyi (Fai- Elkinsia polymorpha, based on the descrip- ron-Demaret and Scheckler 1987), appear to tions and reconstructions of Serbet and Rothwell have been reproductively and, perhaps, vegeta- (1992), was a plant of small, open, shrub-like tively similar to E. polymorpha, whereas others, stature that produced seed-bearing cupules in such as Aglosperma quadrapartita (Hilton and profusion on specialized branches dedicated to Edwards 1996), have reproductive attributes that reproduction. Wind pollination permitted the differ from the other known forms, such as non- plants to occupy primary successional areas as cupulate seeds, suggesting an incompletely un- their principal habitat (Algeo and Scheckler derstood ecological breadth among these early 1998, Algeo et al. 2001). Such habitats initially plants. The earliest indication of seed-like habit appear to have been in lowland wetland settings. precedes these better known seed plants by 20 Scheckler (1986a, b) places these earliest seed million years; Runcaria heinzelinii, an integu- plants in barren, prograding portions of small mented megasporangium surrounded by a cu- deltas, where they were succeeded rapidly by pule (Gerrienne et al. 2004), suggests the early swamps populated by more aggressively grow- appearance of wind delivery of male gameto- ing spore-producing plants. He also suggested phytes to a non-integumented megasporangium growth on levees in the lower delta plain, which, (e.g., Bateman and DiMichele 1994a). Though although wet, were subject to flood disturbance lacking much of the reproductive sophistication and may have been somewhat drier than the im- of Late Devonian seeds, this organ significantly mediate stream sides or floodbasin/backswamp increases our appreciation of the ecological habitats. As far as currently understood, similar complexity of early terrestrial vegetation. growth habits and ecological patterns appear to Elkinsia polymorpha was first described from characterize other Late Devonian seed plants, seeds (Gillespie et al. 1981). It had hydrasper- most of which have similar cupuliferous repro- man reproduction (Rothwell 1986, Rothwell and ductive systems bearing hydrasperman ovules/ Scheckler 1988), a pollination syndrome char- seeds (Rothwell and Scheckler 1988). In gener- acterized by wind delivery of prepollen grains, al, early seed plants appear to have been uncom- open release of swimming sperm, specialized mon to rare elements of Late Devonian ecosys- prepollen capture structures, and specialized pre- tems. Algeo et al. (2001) noted that although pollen isolation morphology within the ovule. seed plants diversified considerably during the 86 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

Late Devonian, they remained ecologically in- known about the anatomical and morphological significant until the extinctions of the major for- diversity of many of these plants, and credible est forming progymnosperms at the end of the whole plant concepts are emerging. In addition, Devonian released ecological resource space. the taxonomic implications of the anatomy and From these beginnings as rare, opportunistic, morphology have begun to crystallize. early successional plants, pteridospermous seed plants rapidly began to diversify during the Mis- MAJOR EVOLUTIONARY LINEAGES AND sissippian as the archaeopterid progymnosperms AuTECOLOGiES. According to Galtier (1992), the disappeared and the seed habit permitted access pteridosperms of the early and middle Missis- to the vast, underexploited resource space of ter- sippian can be grouped into three major clades ra firma (Bateman and DiMichele 1994a). De- based for the most part on common petiolar spite the much lower overall species richness of anatomies, all of which he tentatively considered the time, the ecosystems of the Late Devonian to be members of the order Lyginopteridales. do not appear to have been any more "invasi- The first group consists of the most primitive ble" by evolutionarily advantaged new species gymnosperms, such as Elkinsia, which have than are many modern, relatively species-rich small narrow stems and gracile architecture. systems. This is particularly peculiar, given the Rothwell and Erwin (1987) considered these supposed intrinsic superiority of the seed habit forms to be similar to the progymnospermous over other types of life histories in moisture- group Aneurophytales, evidence that the seed stressed environments, which are generally plants were derived from this group of small imagined to be greatly underpopulated at this progymnosperms rather than from large woody time in Earth history. It is necessary to recall, trees of the Archaeopteridales. Elkinsia is simi- however, that the seed is not the only trait need- lar in its anatomy to several Mississippian gen- ed to advance into areas of periodic moisture era, such as the anatomically preserved forms stress, where penetrating root systems and bio- Laceya, Triradioxylon, and Tristichia (Galtier chemical adaptations are equally important (e.g., 1988), all of which have Lyginopteris (Lyginor- Algeo et al. 2001). Strong incumbent advantage achis)-\ike petiolar anatomy, Sphenopteris-type (physical space = resource occupation as de- foliage, and gracile growth habit (e.g., Scott and scribed by Hubbell [2001], who emphasized that Meyer Berthaud 1985). resource capture happens at the level of individ- A second group, closely related to these ear- ual organisms, not species), in this case meaning lier forms is represented by Calamopitys and pu- occupation of resource and physical space by in- tatively related taxa, again known mainly ana- dividual plants of pre-existing clades, also may tomically, including Diichnia, Triichnia, Galti- have played a major role in keeping seed plants era, Bostonia, and Stenomyelon (Galtier et al. out of the vast majority of environments until 1993). These plants also have lyginopterid-like extinctions released these areas for invasive col- petiolar anatomy and comprise the family Cal- onization. There also are interesting parallels be- amopityaceae. Both groups 1 and 2 have sec- tween the advent of seed plants and the advent ondary xylem, although it is limited in extent. of the angiosperms (Hickey and Doyle 1977)— In addition, for both these groups the tracheids both groups entered well populated ecosystems, of the metaxylem are considerably larger in di- possessing apparently unusually advantageous ameter than those of the secondary xylem. In biological traits relative to the existing occu- habit, as far as it is known, the plants of group pants. In each case they entered and initially di- 2 appear to be scrambling ground cover, possi- versified along stream margins in disturbed set- bly vines or liana-like plants (Fig. 1). This has tings, only later rising to dominance in conjunc- been deduced from their stem architecture, tion with extinctions of the earlier forms: wherein the internodal distance is relatively perhaps passive replacement rather than com- short in large diameter stems but becomes in- petitive displacement, at least initially. creasingly long in the smaller diameter stems, to the point that such small stems appear to have Pteridosperm Ecology in the Early to Mid- been incapable of self-support. Overall, this sug- dle Mississippian: the Emergence of Ecologi- gests an upright or corm-like basal portion with cal Diversity. Pteridosperm species richness a more trailing or "semi-self supporting" upper and the diversity of body plans increased enor- part of the mature plant (Rowe et al. 1993), pos- mously during the Mississippian—a time of ver- sibly with determinate growth (Hotton and Stein itable flowering of the group. Quite a lot is 1994). Fronds in these plants were relatively 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 87

and metaxylem tracheids smaller in diameter than those of the secondary xylem. Because of their larger size and thicker wood, Galtier and Scott (1990) considered group 3 pteridosperms to be longer-lived trees of forested landscapes, and thus more K-strategists (committing re- sources to vegetative tissues that will confer the ability to withstand environmental changes and to undergo repeated reproductive events) com- pared with the other groups of smaller stature. On the other hand, Bateman and Scott (1990) and Bateman (1991) identified a mixture of pte- ridosperms, including some of these larger forms, in volcanigenically disturbed habitats at the classic Oxroad Bay locality in Scotland, ar- guing for pteridospermalean scrub community that included Eristophyton and Bilignea, as well as the more gracile form, Triradioxylon. The larger pteridosperms were likely relatively deep- ly rooted, a morphological innovation supported by studies of fossil soils. Algeo et al. (2001) described several proximate changes in soil de- velopment, including the appearance of more FIG. 1. Carboniferous pteridosperms and the pro- abundant and widespread vertisols, indicating gymnosperm Callixylon-Archaeopteris, Of particular note here are the reconstructed growth habits of Cal- plant colonization of periodically moisture-lim- amopitys (second from left) a small groundcover plant, ited habitats, and changes in the clay mineral and Pitus (third from left), a large tree. Note also Ly- content of soils, notably the increase in clays ginopteris, also reconstructed with as a facultative indicative of longer intervals of weathering climbing plant. Reprinted from Galtier (1986, Fig. 4) with permission. (smectites and kaolinites). They speculate that the colonization of ever more physically stress- ful terra firma environments would have had large, up to 50 cm in length (Galtier 1974, Gal- major implications for global weathering cycles tier et al. 1993). and, through that, had strong effects on atmo- The third group of Mississippian pterido- spheric gaseous composition. Thus, the spread sperm-like plants has larger, woody stems. In- of plants, most likely pteridospermous seed cluded are the genera Eristophyton, Pitus, and plants as foundational elements, into the drier Bilignea, which Galtier (1986) has reconstructed hinterlands is hypothesized to have profoundly as trees (Fig. 1). Eristophyton and Bilignea have affected global ecological patterns. been found in attachment to anatomically pre- All of these pteridosperms or pteridosperm- served petioles of the Lyginorachis type (Long like plants probably had hydrasperman repro- 1987, Bateman and Rothwell 1990). Fronds, ductive biologies. All the well characterized re- possibly attributable to the foliage genera Rha- productive organs from the Late Devonian and copteris and Spathulopteris, and perhaps also Mississippian are similar in their possession of Sphenopteridium or Adiantites, were borne this syndrome (Rothwell and Scott 1992). This densely on terminal shoots (Galtier and Scott includes wind delivery of prepollen, capture by 1994, Galtier et al. 1988). These fronds, about a specialized lagenostome (a funnel-like struc- half a meter long, although larger than older ture that projects from the top of the megaspo- forms (Long 1979a), may be considered rela- rangium), and a pollen chamber that is closed tively small in light of the large size of the par- by the ontogenetic development and upward ent trees. A biomechanical analysis of Pitus dayi growth of the female gametophyte following (Speck and Rowe 1994), showed that the plants prepollen delivery. Integuments in these types of were certainly self-supporting and that they ab- seeds often are composed of unfused lobes or, if scised their fronds. Although, like groups 1 and fused, with very broad micropyles to accom- 2, these plants have lyginopterid-like petiolar modate the lagenostome (see the description in anatomy, they also have thick secondary xylem. Rothwell and Scheckler 1988, p. 116). Such JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

numerous, small seeds (< 10 mm long) found in association with the vegetative remains, both of which point to opportunistic life histories (Er- win et al. 1994, Alleman et al. 1995). The sim- ilarity of such ecological patterns in the early and middle Mississippian tropics and paratropics indicates the broad generality of this autecolog- ical role for most early pteridosperms.

THE ROLE OF PTERIDOSPERMS IN EARLY AND MIDDLE MISSISSIPPIAN PLANT COMMUNITIES. During the Mississippian, seed plants became FIG. 2. Occloa, a Mississippian-age lyginopterid dominant elements of both adpression and pet- pteridosperm forming ground cover in a coastal plain rifaction, clastic substrate lowland (but not bone habitat. Reprinted from LePage and Pfefferkorn (2000, fide swamp) floras (Scheckler 1986a). In one of Fig. 1) with permission from the Paleontological So- the few summaries of early pteridosperm ecol- ciety. ogy, Rothwell and Scheckler (1988) noted that the fossil floras from the early Mississippian are seeds generally are small and can be produced comparatively poorly known relative to those of in abundance. Additionally, many of the seeds the Late Devonian and middle through late Mis- are borne in cupules that themselves have lobes sissippian. Despite considerable increases in the that are fused to varying degrees. The appear- empirical understanding of pteridosperm mor- ance of seed concentrations in sediments (e.g., phology and distribution since that time, the ba- Scott and Meyer-Berthaud 1985, Rothwell and sic conclusions of these authors have proven ro- Scott 1992) suggests that reproduction may have bust. Early and middle Mississippian pterido- been periodic, probably seasonal. Generally sperms, in general, appear to have been most small (£ 5 mm in diameter), seed sizes do not common in disturbed settings, including stream suggest K-selected reproductive strategies (re- levees and drier parts of floodplains in North source allocation to fewer larger seeds, increas- America, and in volcanigenic landscapes in ing the investment in the success of each seed) western Europe. Of particular note is the quan- to the same degree as seen in the development titative study by Galtier et al. (1988) of the mid- of larger stem sizes in the group 3 pteridosperms dle Tournaisian petrifaction flora (early Missis- described above. The production of many small sippian) from Montague Noire, France. Three of seeds, particularly if lacking dormancy, which four sampling sites were dominated by pterido- seems likely given their open integumentary sperms, particularly by the Calymopityaceae structure, continues to suggest the ability to in- (group 2 above), at levels of 63—100% of the vade and exploit disturbed landscapes. For ex- specimens recovered. These are largely ground ample, Eristophyton, Bilignea and, in some cas- cover plants, perhaps thicket forming. They are es, the other large-statured forms have been re- associated with small percentages of ferns and, ported to be a significant component in volcan- at one site, with about 13% lycopsids. One other igenically influenced environments (e.g., Scott site was dominated by the zygopterid fern, Clep- and Galtier 1988, Bateman and Scott 1990, sydropsis, with only minor amounts of pterido- Bateman 1991, Galtier et al. 1993). In some in- sperms, suggesting some degree of landscape stances, these plant fossils are closely associated heterogeneity. The flora is preserved in cherts with fossil charcoal, an indication that the land- interbedded with phosphatic nodules, indicating scapes were repeatedly swept by wildfires (e.g., transport from the original environment of Scott et al. 1994). growth into standing water. Late Visean (late middle Mississippian) lygi- Scotland is perhaps the best known area for nopterid pteridosperms from the Paracan floral early and middle Mississippian (Tournaisian and realm, a region between the south temperate Visean) terrestrial fossil deposits from Euramer- Gondwanan and the tropical Euramerican floral ica (see classic work of Long 1960a, b; 1961a, realms (lannuzzi and Pfefferkorn 2002), have b, c; 1964, 1965, 1969, 1975, 1976, 1977a, b, c; been found to cover floodplains in monospecific 1979a, b). A few examples of these studies re- stands (LePage and Pfefferkorn 2000) (Fig. 2). veal the general patterns that have been found. This pattern of distribution is associated with Perhaps the best known of these sites is Oxroad 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 89

Bay of late Tournaisian age (Scott and Galtier sensu stricto; possibly related to Paralycopodi- 1988, Bateman and Scott 1990, Bateman 1991), tes; Bateman, personal communication), the ca- which preserves a wide array of both petrifac- lymopityacean ground cover pteridosperm, Sten- tion and compression plant fossils (Bateman and omyelon, and some small ferns. When compared Roth well 1990). A succession of environments with other floras from approximately the same has been identified, beginning with flood-plain time interval and environmental conditions deposits that include both small tree forms, such (Oxroad Bay—volcanigenic; Cove, Burnmouth, as Eristophyton, and pteridosperms with shrubby Foulden, Edrom—fluvio-lacustrine), Scott and growth habit, such as Stenomyelon, some species Galtier (1988) echoed the conclusions of Roth- of which are restricted to these wetter land- well and Scheckler (1988), arguing that pteri- scapes, which are the least disturbed of the dosperms, large and small, appear to have been environmental succession. The early fluvial en- particularly adapted to growth in disturbed en- vironments are succeeded by a series of volcan- vironments. Some small lycopsids, such as igenically disturbed landscapes, in which pteri- Oxroadia, also appear to have been able to sur- dosperms are common and abundant, dominat- vive under disturbed conditions (Bateman 1992), ing some of the assemblages. Total species di- as were many early ferns (Scott and Galtier versity in these landscapes was low, between 6— 1985). ^^Lepidodendron" calamopsoides, in 8 species on a reconstructed whole-plant basis, contrast, was not capable of survival under the with pteridosperms accounting for half to nearly conditions of heavy disturbance by repeated vol- all the species at any one site, especially in the canism, which created ephemeral opportunistic more heavily disturbed settings. Bateman (1991) settings. noted that these pteridosperms formed a persis- These Tournaisian patterns continue into the tent scrub community consisting of the tree Visean. For example, Galtier et al. (1993) stud- forms Eristophyton and Bilignea, and the gracile ied a flora from the Weaklaw site in East Lothian form, Triradioxylon. Low species richness, and preserved in a volcanigenic sequence. The flora the persistence and dominance of pteridosperms from this site, found in several ash beds, is uni- of several growth habits, suggest fairly wide formly pteridosperm dominated, but only by the ranging environmental tolerances and opportu- large woody forms Eristophyton and Pitus, and nistic life histories for a number of species. Scott the smaller tree or possible shrub, Bilignea. Also and Meyer-Berthaud (1985) described another present are some compression foliage and the late Tournaisian volcanigenic landscape from zygopterid Diplolabis. Several other volcanigen- Foulden in which fossil plants were preserved ic sites of similar age are dominated by ferns as both petrifactions and compressions in two and/or lycopsids. Yet other sites are dominated different beds. The gracile, primitive pterido- by pteridosperms, or a mixture of pteridosperms sperm Tristichia dominates one of the beds in and ferns/lycopsids. Included are the Pettycur, association with the seed Stamnostoma. The de- East Kirkton, and Loch Humphrey Burn locali- posit appears to have been a standing water en- ties of Scotland, and the Esnost, and Roannais vironment, possibly a lagoon, in which spiror- localities of France. Galtier et al. (1993) con- bids could colonize locally transported plant ma- cluded from this analysis that there were two terial. In a deposit from the Castleton Bay lo- basic community types in these landscapes: cality in East Lothian, of approximately those dominated by pteridospermous seed plants equivalent (late Tournaisian) age to that from and those dominated by ferns and/or lycopsids. Foulden, Scott and Galtier (1988) described A small window in the later Mississippian plants in a wet floodplain deposit located be- suggests that the flora described above continued tween then-active volcanic vents. The plants to survive in disturbed, volcanigenic landscapes, were deposited in abandoned channels, after which may, in fact, have constituted refugia in some transport, in small ponds where carbonate a world that was increasingly Pennsylvanian in was precipitated, preserving the plants as petri- aspect (evolutionary modernization of the plants factions. Included were spirorbid worms, ostra- with patterns of ecological partitioning more like cods, and fish bones, suggesting brackish water, that of the tropical wetland landscapes that possibly estuarine conditions. In this instance, would dominate peat-forming lowlands). Chal- the flora is dominated by the arborescent pteri- ot-Prat and Galtier (1989) described a woody dosperm Eristophyton, but with a significant el- trunk fragment, tentatively attributed to Eristo- ement of the small lycopsid "Lepidodendron" phyton, from Tazekka, Morocco, in rocks con- calamopsoides (not actually a Lepidodendron, siderably younger than those of the Scottish or 90 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

French occurrences from which this genus is re- The "Coal Swamp" Landscape. The floras ported in abundance. The local sedimentary set- associated with coal beds of Pennsylvanian age ting is complex and the cf. Eristophyton speci- (late Namurian, Westphalian, and Stephanian) men is preserved in volcanigenic sediments dis- are among the best studied fossil floras. Well tinct from, but associated with, floras dominated exposed in coal mines, plants have been studied by lycopsids (Lepidophloios) in swamp deposits for centuries and are known both as petrifactions and calamitaleans (Mesocalamites) in clastic and adpressions. The mainly carbonate permi- flood-basin sediments. This parallels patterns neralizations of peat stages of coal, known as found in the Tournaisian-Visean where pterido- coal balls, have yielded unrivaled information sperms often were found interbedded with, but on plant anatomy, development, and reproduc- in different facies from, lycopsids, ferns, and tive biology. However, with a few notable ex- sphenopsids. In this case, the lycopsid and cal- ceptions (e.g., Rothwell 1981), they have left amite are of considerably more advanced aspect much to be desired regarding whole-plant recon- than those of the early Mississippian. In differ- structions or a clear picture of whole-plant (as ent sedimentary conditions, Gerrienne et al. opposed to organ-level) species-level taxonomic (1999) described a flora from the late Mississip- diversity. Quantitative studies of coal-ball floras pian of Belgium that contained Eristophyton-\ike have permitted large-scale analyses of commu- wood in association with what are possibly the nity dynamics and responses of ecosystems to oldest records of Arthropitys calamitaleans and climate change (e.g., Phillips et al. 1985, Di- the marattialean tree fern Psaronius, clear har- Michele and Phillips 1996b). Adpressions, on bingers of later Pennsylvanian wetland floras. the other hand, have revealed rare but significant These changes in the composition of the dom- insights into plant architecture (e.g., Bertrand inant floras in Euramerica may have been driven and Corsin 1950, Laveine et al. 1977, 1997; by changes in continental positions, as Pangea Laveine and Duquesne 1998, Wnuk and Pfef- accreted and formerly paratropical regions ferkorn 1984, Shute and Cleal 2002), in situ moved equatorward (Raymond 1985) from areas plant spacing and landscape structure (e.g., Gas- of strong seasonality to areas of less seasonal taldo 1986, 1987), plant responses to glacial-in- climate. However, they also may reflect the on- terglacial climatic changes (e.g., Falcon-Lang set of the Carboniferous glacial interval, which 2003, 2004), and the complexities of taxonomic would have produced strong polar high pressure richness (e.g., Krings et al. 2003b). It has been cells and narrowed the atmospheric intertropical difficult to reconcile or cross-correlate the pet- convergence zone, increasing rainfall at the rifaction and adpression records; the differences equator and fundamentally changing tropical cli- in preservation frequently lead to different suites matic dynamics (Cecil et al. 2003). of characters, different traditions in the way the In summary, the data that have appeared since fossils are studied, and so forth. the publication of the review by Rothwell and The origins of the "coal-swamp" flora are un- Scheckler (1988) have expanded and confirmed certain. Was it assembled species-by-species their observations. The early pteridosperms, from taxa that evolved elsewhere before entering both in Devonian and early Mississippian trop- the tropical lowlands, or did the flora evolve ical landscapes appear to have been opportunis- more as a unit that was assembled in situ from tic forms, centered ecologically in disturbed en- diversifying lineages as the glacial-interglacial vironments. This was especially true where the rhythms of the latest Mississippian and Penn- disturbances were floods, levee progradation, or sylvanian began, taking the Earth from a warm volcanigenic events, which created opportunities to a cool global climate (Gastaldo et al. 1996)? for primary succession. The plants typically had The pteridosperm element of this flora may shed rapid growth, early onset of reproduction, xeri- some light on the problem. Some of the pteri- cally adapted shoot morphologies, highly rami- dosperm lineages typical of peat-forming, wet- fying root systems, wind pollination, and pro- land landscapes appear prior to the major de- duction of large numbers of small, highly dis- velopment of such landscapes, in association persible seeds. The ultimate evolutionary and with various elements of the earlier Mississip- ecological fate seems to have been extinction for pian floras. the large tree forms but some appear to have been the evolutionary ancestors of the pterido- EARLY APPEARANCES OF WETLAND, TROPICAL sperms that were of importance in Pennsylva- PTERIDOSPERMS. Taxa and growth habits. The nian wetland landscapes. two "classic" groups of Pennsylvanian-age pte- 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 91 ridosperm lineages are the orders Lyginopteri- late Namurian A age from a wide geographical dales and the Medullosales, each encompassing area (Taylor and Eggert 1967, Mapes and Roth- large taxonomic diversities. Each also appears to well 1980, Dunn et al. 2003b). The oldest well have its origins very early in, or before, the in- documented specimens attributable to Medullosa ferred onset of glacial conditions. (M. steinii: Dunn et al. 2003b) have anatomical As noted in the discussion of earlier Missis- features consistent with vine/climber or scram- sippian pteridosperms, the matter of the mor- bling habit. These include small stem diameters, phological circumscription of the Lyginopteri- spines on the bases of petioles that may have dales (Calamopityaceae, Lyginopteridaceae, per- served as climber hooks, which are well docu- haps others) and included families remains prob- mented in adpressions from the Pennsylvanian lematic (Meyer-Berthaud 1990). Nonetheless, (Krings et al. 2003b), and a vascular system plants with anatomical similarities to Pennsyl- composed of cable-like, anastomosing, vascular vanian Lyginopteris have been described as ear- strands. These features suggest a flexible stem ly as the Visean, from the classic Pettycur lo- that was incapable of self-support. The stelar ar- cality in Scotland (Bertram 1989). The family chitecture of medullosans has long been per- Lyginopteridaceae can be extended even further plexing because of its "vine-like" structure, back into the Tournaisian if one is willing to composed of distinct vascular bundles, each sur- accept the assignment of such taxa as Laceya or rounded by a cylinder of secondary xylem, rath- Tristichia to the Lyginopteridaceae (e.g., Dunn er than a continuous band of wood surrounding et al. 2003a). More certain records of the genus the stele to the external side. In addition, the Lyginopteris place it as early as the upper Vi- stems are rich in longitudinal fibrous bundles, sean, based on a suite of vegetative and repro- which also confer flexibility in stem and leaf ductive organs described as Lagenopteris ber- construction. It is difficult to have confidence mudensiformis (Hartung 1938), and from the that first stratigraphic occurrences are also the Namurian A, during the early phases of the on- earliest/most primitive evolutionary states. set of the late Paleozoic ice age, as compression Nonetheless, the early occurrence of such a vine foliage (Patteisky 1957, Pfefferkorn and Gilles- or liana-like architecture in Medullosa, which pie 1982), from anatomical preservation (Lygi- also is a characteristic of most subsequent mem- nopteris royalii: Tomescu et al. 2001), and as bers of the Medullosales, presents the possibility prepollen (Schultzispora sp.) in coal and organic that these were primitive features that were de- shale beds (Peppers 1996, Eble and Greb 2004). velopmentally canalized and thus fundamental Another member of the family, also described architectural elements. The later evolution of from Namurian A anatomical specimens, is Tri- larger tree forms was only possible through venia arkansana (Dunn et al. 2003a). Based on modification of this ground plan, mainly through their anatomical features, both L. royalii and T. strengthening by adding features such as exten- arkansana are interpreted as vines or scrambling sive secondary xylem or large quantities of scle- ground cover. Another probable lyginopterid renchyma. climber or scrambler, from the Mississippian/ Quaestora amplecta is a small woody stem Namurian of Britain, was described as Rhetin- attributed to the medullosan evolutionary line- angium arberi by Gordon (1912). This plant is age (Mapes and Rothwell 1980). The plant is preserved anatomically and has small diameter, known only from small diameter specimens with woody stems with radially aligned sclerenchy- a cruciate vascular system surrounded by sec- matous bands in the cortex, massive leaf traces ondary xylem, rather than the separate woody and petiole bases, and petiolar anatomy of the bundles typical of Medullosa. It may have been Lyginorachis type. The genus Heterangium, also a small tree, possibly part of a subcanopy in ear- assigned to the Lyginopteridaceae, and a mem- ly wetland ecosystems (Dunn 2004, Dunn et al. ber of Pennsylvanian peat-forming communities, 2006). This plant is important because it sug- has been described as early as the Visean of gests early ecological diversity in the Medullo- Britain in anatomical preservation (Scott et al. sales, something typical of this group during the 1984); like other lyginopterids, it is interpreted Pennsylvanian. as a vine or scrambling plant based on small Medullosan seeds also have been discovered stem diameters and sclerenchymatous plates in in Namurian A deposits (Dunn et al. 2002b). At- the cortical tissues. tributable to Rhyncosperma quinnii, these seeds The order Medullosales also is first recog- have features that link them to trigonocarpalean nized as anatomically preserved specimens of seeds, particularly the medullosan genus Ste- 92 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133 phanospermum. Rhyncosperma quinnii is a Lyginopteris, as a major floristic component in moderate sized seed, reaching lengths of 22 mm, swampy habitats dominated by spores of the ly- indicating that significant investment of energy copsid tree Lepidophloios. Schultzispora rara in seeds was an early appearing trait in this remains a component of coal palynofloras group of plants, something that would reach though the Langsettian (Westphalian A), when much greater proportions in the Pennsylvanian Lyginopteris becomes extinct, and is frequently wetlands. associated with lycopsid trees in coal-ball floras. Assuming the parent plant to be a vine or scram- Pteridosperms in early tropical wetland bling plant, based on the foliage and stem anat- communities. The late Mississippian is a crucial omy of later lyginopterids, it appears that there time interval for the evolution of the later Car- was a significant pteridospermous component of boniferous wetland biome, but remarkably little wetland vegetation early in the history of the has been published about it from a synecological tropical wetland biome. perspective. One of the few studies is that of Dunn (2004; Dunn et al. 2006), which considers PTERIDOSPERMS IN PENNSYLVANIAN-AGE the flora of the Fayetteville Formation of Arkan- WETLANDS. Medullosans. Taxonomic Rela- sas of approximately Namurian A age. Three tionships. The Medullosales is a complex group distinct plant communities have been recognized of plants with a variety of growth forms, frond based on both adpression fossils, from terrestrial architectures, and reproductive biologies. Orga- sandstone and mudstone, and petrifaction fos- nization of this variation within a well supported sils, transported into marine black shale. One of phylogenetic framework would be most wel- these communities is dominated by lycopsid come because it would permit a clearer evalua- trees, one by pteridosperms, and one by oppor- tion of the relationship between ecological role tunistic ferns. In the overall flora, consisting of and evolutionary relatedness. However, at pre- 15 whole-plant species, pteridosperms, although sent, there are few whole-plant reconstructions not numerically dominant, are by far the most on which such relationship schemes can be diverse group at the species level (10 species), based. One such attempt is that of Laveine with most of that diversity invested in ground (1997), who aligned medullosan taxonomic re- cover and vines (7 species) rather than canopy lationships on the basis of frond architecture and trees (1 possible pteridosperm) or understory foliar morphology, and attempted to include trees and shrubs (2 species). The canopy domi- what is known of the association between major nant has characteristics that would tie it more types and taxa of dispersed organs. Laveine closely to earlier Mississippian "lyginopterids" (1997) recognized three "family" level groups, than to later tropical wetland taxa. Similar pat- which are outlined below. This summary per- terns of distribution for the small pteridosperm mits some of the basic ecologically relevant as- trees, Quaestora amplecta and an undescribed pects of the architecture of these plants to be "lyginopterid", were taken to suggest that these summarized. plants occupied the understory, also in such dri- Neuralethopteridaceae includes the foliage er habitats. Possibly lyginopterid foliage, as- genera Alethopteris, Lonchopteris, Neuraleth- signed to two different species of Sphenopteris, opteris, Cardioneuropteris, and possibly Neu- formed an abundant ground-cover component in rocardiopteris. Lonchopteridium is also a prob- each of two lycopsid dominated assemblages. able member of this group (Zodrow and Cleal Perhaps the most important findings from this 1998). These forms are expected to have a "bi- study, relevant to this review, are that pterido- furcate pinnate" frond architecture, in which the sperms were diverse in their ecological roles petiole is naked. These kinds of fronds can be early in the appearance of tropical wetland veg- very large, up to 7—8 m long and 4 m wide. etation, and that they were mainly ground cover Where known, these genera have Aulacotheca- and vines, possibly shrubs, rather than major Whittleseya pollen organs and Trigonocarpus canopy trees. This would change during the seeds. Alethopteris is abundant in western Eu- Pennsylvanian as some pteridosperms broadened rope during the late Namurian and early West- into the role of local canopy dominants. phalian, can be abundant in the Stephanian, but Palynological analyses of coals and organic is rare in the "Rotliegend/Autunian." It does not shales from the late Mississippian (Chesterian) appear in China until the Stephanian, and then of the U.S.A. (Eble and Greb 2004) have iden- only occasionally. This clade may be the ances- tified Schultzispora rara, the prepollen grain of tral form for this entire group of plants. Based 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 93 on frond architecture, there is evidence that the some cases, for example, specific adpression earhest described medullosan fohage, "Neurop- taxa can be linked to anatomy and tied to growth teris" antecendens, will likely prove to have a habits and sedimentary environmental settings, 'bifurcate pinnate' type frond. The same frond which permits ecology to be understood, subject type is found in Neurocardiopteris broilii, a ge- to reasoned speculation. In other instances there neric name that, according to Laveine (1997), is only, say, foliage, linked partially or not at all has precedence for these early forms. to anatomy or whole-plant concepts. Or there is Parispermaceae include the foliage genera the vast information on seeds and their sizes, Paripteris and Linopteris. These forms have most of it not clearly linked to specific whole- "pseudopinnate" compound fronds with two plant concepts. terminal pinnules on each pinna rachis (paripin- As part of this general set of background data, nate condition). There are no pinnule lobations. it can be determined that medullosans displayed Fronds are large and forked, with naked petioles a wide spectrum of growth forms. Clearly doc- and pinnules on rachises of all orders above the umented are climbing vines or liana-like mor- fork. They are associated with anatomically pre- phologies, thicket-forming "leaning" tree forms served stems of the Sutcliffia-type (Stidd et al. that apparently were not self-supporting, and 1975) and have Potoniea pollen organs and Hex- free-standing trees. However, for many taxa at agonocarpus seeds (for summary, see Laveine the species level, the details of the growth habit, et al. 1993). other than tree, vine, etc., remain unknown. Neurodontopteridaceae of Laveine was split Tree forms. At the most general level, Pfef- by Cleal and Shute (2003) into two families. The ferkorn et al. (1984) and Wnuk and Pfefferkorn Neurodontopteridaceae, sensu stricto, includes (1984) described stems from in situ deposits of Neuropteris, Reticulopteris, Odontopteris, Ma- middle and late Pennsylvanian age (Bolsovian/ croneuropteris, and possibly Neurocallipteris, Westphalian C through Stephanian) from shales Neurodontopteris, and Barthelopteris. The Cy- above and below coal beds in the U.S.A. and clopteridaceae include Laveineopteris, Margar- France. Two medullosan tree growth architec- itopteris, and Calipteridium. Laveine also in- tures were identified based on prostrate trunks cluded Cardioneura, Cardiopteridium, Spheno- with leaf bases (though not identifiable foliage) neuropteris in his familial concept (see also attached. One tree architecture consists of up- Cleal and Shute 1995). The earliest occurrences right stems as much as 20 cm in diameter or of Neuropteris reliably reported date the genus more (based on these published papers and on to the Namurian A of the latest Mississippian subsequent personal field observations), proba- (Purkyiiova 1970) and appear attributable to A^. bly less than 5 m in height, with closely spaced, obliqua. Plants of these two families have bifur- densely packed fronds, about 10 per meter of cate, semi-pinnate fronds with pinnae (ultimate stem length. The fronds appear to have recurved or bipinnate depending on the genus). In the after death, possibly forming a "skirt" around Neurodontopteridaceae 5.5. large lateral pinnules the plant prior to rotting or breaking off; skirts are found below the main fork of the frond. As are generally considered to be morphological described by Cleal and Shute (2003) true cy- mechanisms for shedding epiphytes and vines in clopterid pinnules occur in the Cyclopteridaceae. modern plants such as palms or some tree ferns. Intercalary ultimate pinnae may occur between Such trees appear to have been solitary with live insertions of bipinnate lateral pinnae. In each of fronds concentrated in a tuft near the plant apex the families, the apex of each major frond seg- (Fig. 3). The other growth form consisted of ment can have pinnules on the rachis and ter- flexuose stems up to 13 cm in diameter and minates with a single pinnule (imparipinnate). greater than 5 m in height, with widely spaced Fronds of Macroneuropteris macrophylla have fronds, 3.0—4.5 cm in diameter at the base, been reconstructed from large partial specimens which appear to have been shed regularly; there (Cleal et al. 1996). Some species of neuropterid is no indication of a dense skirt of dead fronds genera have been shown to be attached unequiv- (see also specimen illustrated by Demko and ocally to stems with Medullosa anatomy (Beeh- Gastaldo 1992) (Fig. 4). These latter trees appear ler 1983). to have formed thickets in which tall, but rela- Within this basic taxonomic configuration, in- tively small diameter; stems were able to support formation about medullosan species remains a mass of fronds by intertwining with adjacent variously connected and disconnected with an stems; the authors report finding similar growth increasing number of centers of attraction. In habits in modern tropical flowering plants. Pfef- 94 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

FIG. 3. Free-standing meduUosan pteridosperm growth habit. Reprinted from Wnuk and Pfefferkom (1984, Fig. 11) with permission from Elsevier FlG. 4. Lax-stemmed medullosan pteridosperm growth habit. Reprinted from Wnuk and Pfefferkom ferkorn et aL (1984) evaluated published recon- (1984, Fig. 10) with permission from Elsevier structions of medullosans, which show a wide range of variation in the interpretation of flexu- ose vs. upright habit (Fig. 5). Some of these re- croneuropteris scheuchzeri, Neuropteris ovata, constructions appear to be based on plants with and Laveineopteris rarinen'is. dead or dying foliage or are unlikely for plants Shute and Cleal (2002) reevaluated Lavei- growing in natural stands of vegetation. neopteris specifically and suggested a growth Adpressed foliage associated with these stems habit for the plant that largely accorded with the was attributed to two major groups of meduUo- findings of Wnuk and Pfefferkorn (1984) and san pteridosperms, the alethopterids and neurop- Pfefferkorn et al. (1984) but showed some in- terids. Wnuk and Pfefferkorn (1984), after ex- teresting ecological elaboration. Some neurop- amining the remaining segments of attached fo- terids are associated with peculiar leaves termed liage, attributed Alethopteris sp. to flexuose stem "cyclopterids"—laminate, leaf-like organs gen- morphologies; however, this assertion was dis- erally placed near the base of fronds, along the puted by Laveine (1997), who determined that petioles, or at the point of frond bifurcations. the frond of Alethopteris and related genera has Shute and Cleal (2002) suggested that in Lav- a different architecture than that described by eineopteris cyclopterids are actually leaves of Wnuk and Pfefferkorn (1984). Neuropterid pole-like juvenile plants growing in deep shade forms could be less confidently assigned to ar- in the subcanopy. They argued that many of the chitectural stem categories and were found in undifferentiated smaller axes, attributed by association with both flexuose and upright stem Wnuk and Pfefferkorn (1984) to petiolar re- morphologies. The associated neuropterids, in mains, are, in fact, such pole-like juvenile stems. the current taxonomic system (Cleal et al. 1990), Such stems would have borne cyclopterid fo- can be assigned to species in three genera: Ma- liage that left only cryptic scars where those cy- 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 95

contrast to other species of this genus, cyclop- terids are found relatively rarely in association with L. rarinervis, which might be consistent with a liana-like or vine habit where long-lived shade leaves might be less important than in plants that spent much of their life in the under- story (C. J. Cleal, personal communication, 2005). Perhaps all these sources of data and in- terpretation can be reconciled by imagining ear- ly growth as a small pole-like sapling with het- eroblastic leaf development, bearing fronds only when reaching the canopy, there supported by intertwining with other individuals, forming a largely understory thicket. The largest medullosans, based on stem di- ameters of permineralized specimens, have been reported from the Pennsylvanian-Permian tran- sition in Germany and France. Sterzel (1918) re- ported stems of Medullosa leuckartii from Chemnitz, in the German Rotliegend, with di- ameters of up to 26 cm, bearing in helical phyl- lotaxis petioles up to 16 cm in diameter with massive secondary xylem. One fragmentary specimen was nearly two meters in length, sug- FIG. 5. Reconstructions of free-standing medullo- gesting considerable height. The stems were as- san pteridosperms (A—Andrews, 1947; B, C, D—Ber- sociated with adpressed foliage attributed to trand and Corsin, 1950; E- Stewart and Delevoryas, Neurocallipteris planchardii (moved from Neu- 1956; F—Corsin in Buisine, 1961; G—Pfefferkorn et ropteris by Cleal et al. 1990) and Neurodontop- al., 1984). Reconstructions B, C, D, and G liave tlie appearance of tropical plants. Reconstructions A and teris auriculata (which Potonie 1893, segregated E show plant with drooping leaves that are either dead from Neuropteris, a determination reaffirmed by or dying. Reconstruction F is a growth form that is Kerp and Krings 2003). Stems of similar size unlikely to occur in monocaulous tropical plants. Re- were reported by Renault (1896) from the Autun printed from Pfefferkorn et al. (1984, Fig. 1) with per- mission from the Delaware Valley Paleontological So- Basin of France. These medullosans are quali- ciety. tatively different in size and perhaps growth ar- chitecture from any known earlier. The authors' observations of coal-ball specimens from the clopterid leaves abscissed. In their growth mod- middle and late Pennsylvanian also suggest that el, the typical laveineopterid foliage was pro- maximum medullosan stem sizes were greater in duced once the trees reached the full sunlight of late Pennsylvanian peat-forming swamps and the canopy. They proposed that Laveineopteris mires than earlier. Medullosans of the late Penn- species formed dense stands from which com- sylvanian and Early Permian thus may have peting pteridosperms were excluded by the elaborated on the stout tree morphology de- depth of understory shade. However, Krings et scribed by Wnuk and Pfefferkorn (1984) to al. (2003b) considered Laveineopteris (Neurop- reach even greater dimensions, probably achiev- teris) rarinen'is to be a vine because fronds up ing considerable heights and large frond sizes. to 80 cm long have been found on stems only 3 Vine and Liana-like Growth Habits. Some cm in diameter. Reihman and Schabilion (1978) medullosans appear to have had scrambling or found permineralized specimens of L. rarinervis vine/liana-like growth morphologies. This has to be mesomorphic with the capability to excrete been described in Medullosa steinii, for exam- water, favoring habitats of high soil moisture ple, the earliest described species of this genus and limited evapotranspiration. This is in keep- from the Mississippian. Because it is based on ing with the interpretation of Shute and Cleal anatomical preservation, this species is assigned (2002) that this species specialized in low light to Medullosa. The adpressed foliage taxon to conditions, possibly forming understory thick- which this plant belongs is not clear (Dunn et ets. Shute and Cleal (2002) also noted that, in al. 2003b). Climbing or scrambling medullosans 96 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133 appear to have been most prominent in the late intermontane basins in central Europe, which Pennsylvanian and earhest Permian although, by lacked marine influence and were possibly at extrapolation of generic ranges and from ana- higher altitudes than those of North America and tomical studies (e.g., Scott 1899, Schopf 1939), western Europe. In these environments, O. brar- also occurred in the earlier Pennsylvanian as dii is the most commonly encountered species well. Potential reasons for the increase in vines of this genus. Thus, there are both biogeographic in the late Pennsylvanian are discussed later (see patterns and associated habitat differences dis- subsection entitled "Medullosans in the Land- cernable in species of this genus. scape"). Lescuropteris genuina, one of two species in Prominent among these scrambling and this genus, was found by Krings and Kerp climbing forms are plants associated with the (1997) to bear fronds about 60 cm long, small adpressed foliage Odontopteris, particularly the for a medullosan, but with terminal pinnules "small pinnuled" forms (Simunek and Cleal modified in some instances to form tendrils, sim- 2004), and Lescuropteris (Krings and Kerp ilar to those found in some modern flowering 1997). Hamer and Rothwell (1988), following plants, suggesting a climbing habit. These au- Baxter (1949), described the late Pennsylvanian, thors attributed Zeiller's (1888) stem specimen anatomically preserved Medullosa endocentrica, (mentioned above) to Lescuropteris genuina a plant with odontopterid-like foliage, as a vine based on the presence of an associated frond or liana-like plant based on the apparent delayed segment found with, but not in attachment to, development of the leaves (a pattern typical of the original stem specimen. This supports Si- vines), the absence of adventitious roots, slender munek and deal's (2004) architectural interpre- stems with axillary branching, and long inter- tation, although it results in assignment of the nodes. Fronds of the "small-pinnuled" group of stem to a different, but related, genus. Odontopteris are small, perhaps up to 1 m max- A final medullosan with climbing growth ar- imum length. Although growth habit is not chitecture is Blanzyopteris praedentata, de- known for certain, small tree form was suggest- scribed by Krings and Kerp (1999) based on ma- ed by Bertrand and Corsin (1950), based on sim- terial with small fronds and slender stems. It also ilarity to neuropterids and on a stem specimen appears to be broadly neuropterid in its affini- figured by Zeiller (1888), which Pfefferkorn et ties, based on forked frond morphology, pinnule al. (1984) attributed to their "leaning" growth venation, and pinnule attachment to rachises. morphology. Simunek and Cleal (2004), on the Blanzyopteris praedentata fronds are borne in other hand, argued that this stem is fully com- subopposite pairs with one member of a pair patible with a scrambling plant, similar to Cal- highly modified such that pinnae end in pads listophytonlDicksonites pluckenetii as recon- that are interpreted as adhesive pads. Fronds are structed by Rothwell (1981) or by Galtier and covered with a dense layer of hairs of different Bethoux (2002). Simunek and Cleal (2004) kinds. Some (of a type also found in Odontop- made several interesting observations relevant to teris and Lescuropteris) are bent apically and the species-level autecology of this group. Se- may have deterred insect predators. Others, with cretory ducts, for example, are present in the late swollen bases and filaments of small cells, may Pennsylvanian and Early Permian Odontopteris have broken off when touched by animals such reichiana and O. nemejcii, which may have as arthropods, causing the release of repulsive helped repel piercing and sucking insects, but chemicals (Krings et al. 2002). appear to be absent in the late Pennsylvanian O. Cormose stems. Not discussed in these con- brardii (Kerp and Krings 2003). They also note siderations of growth habit is the possibility of that hydathodes, structures that permit a plant to medullosans with cormose stems and large excrete excess water, appear to be present in O. fronds, emanating from near ground level. Such brardii, based on specimens illustrated by Kerp a growth form might be similar to that of mod- and Krings (2003), but are absent in O. reichi- ern Nypa palms or Angiopteris evecta. Unfor- ana and O. nemejcii. Similarly, lower vein den- tunately, such a growth form would be difficult sity in O. nemejcii than in O. brardii may be a to find preserved in the fossil record, especially further means of reducing water loss and an in- considering the small number of reported ad- dication of growth under drier habitat condi- pressed stems with significant information on tions. Odontopteris reichiana and O. nemejcii frond attachment. Yet, such a habit might be may have occurred in environments with slightly conceivable, especially for very large fronds of more moisture stress than O. brardii, such as some species of Alethopteris and its relatives 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 97

(Laveine 1986, 1997). At present, there is no ytesta (petrifaction) (Hoskins and Cross 1946). direct evidence for such a growth form. Unfortunately, correlation of seeds with partic- Medullosan reproductive biology. Medullo- ular foliage species or stem anatomies is rather sans are characterized by large seeds, large pre- limited and many of the seeds form complexes pollen grains, and generally low reproductive that differ subtly, making fragmentary speci- output, considering the general rarity of these mens difficult to differentiate to more than a spe- organs in comparison to vegetative remains. In cies complex. Based mainly on evidence from general, these features point to "K-selected" re- adpression fossils, seeds appear to have been at- productive strategies. Based on analogy with ex- tached to fronds, in many cases replacing pin- tant plants that have large seeds and swimming nules within the architecture of the frond, or oc- sperm, such as Ginkgo, which has long time in- curring along frond rachis segments. The largest tervals between pollination and fertilization, seeds were gargantuan, reaching lengths of 7— even weedy medullosans may have been rela- 11 cm in petrifaction (Pachytesta incrassata, P. tively slow to germinate and establish, when gigantea, P. noei, P. vera; Taylor 1965) and 10— compared to modern weedy angiosperms. Plants 12 cm in adpressed forms {Trigonocarpus, au- with this type of reproduction would be expect- thors' observations and dimensions as summa- ed to be long-lived, reproduction extending over rized in Gastaldo and Matten 1978). Even the many seasons, accompanied by low success smaller medullosan seeds approximated a cen- rates when viewed relative to the lifetime of an timeter in length, which still makes them large individual plant. Alternatively, some medullo- by modern standards. Virtually all species had sans may simply have reproduced infrequently, complex seed coats composed of multiple lay- even if abundantly, similar to living cycads, as ers. The outer layer, or sarcotesta, typically was argued by Dimitrova et al. (in press). Medullo- composed of large parenchymatous cells and san seeds (e.g., Serbet and Rothwell 1995) and may have been fleshy, in some instances con- pollen organs (Millay and Taylor 1979) exhibit taining resins or other intracellular secondary a wide range of sizes and degrees of complexity, compounds. The middle layer of most medul- which might be assumed to indicate ecological losan seed coats, the sclerotesta, was typically differences, perhaps in response to attack by ar- sclerenchymatous, composed of stony cells. thropods or in relation to pollen dispersal and Within the seed was a large, multicellular ga- pollination (e.g., Taylor 1978, Taylor and Millay metophyte, which served as the energy reserve 1981), and in seed dispersal syndromes. Varia- for the developing embryo, should one be pre- tion in seed sizes also may indicate differences sent. in rapidity of germination and the allocation of Given their large size, many of these seeds resources to many vs. few seeds during any one may have fallen close to the parental plants, episode of reproduction. However, lacking good which might account for the reports of thickets correlations with each other or with parental in several different medullosan species (e.g., plants and foliage types, it is difficult to do more Wnuk and Pfefferkorn 1984, Shute and Cleal than reason from the limited information such 2002). More extensive dispersal of such large reproductive organs present. The structural di- seeds most likely occurred by either animal vec- versity such organs encompass is consistent, tors or water. Certainly, anyone who has col- however, with a wide spectrum of medullosan lected extensively in the field occasionally has ecological roles, as inferred from other sources seen masses of small pteridospermous seeds, of information, some of which have been de- probably of lyginopterid origin, strewn across tailed herein. bedding surfaces in lake and lagoonal deposits Medullosan seeds and ecological implica- or filling troughs in coarser sediments deposited tions: Medullosans had large seeds, which is part in flowing water. Such small-seed accumulations of their morphological and ecological mystique. imply periodic, probably seasonal, seed produc- The medullosan seed literature is large and de- tion and large reproductive output. In contrast, tailed, leading to considerable knowledge about the larger medullosan seeds, although they rare- their female reproductive organs (see Crookall ly can be found in large accumulations, gener- 1976, for a review of adpression forms, and ally are found isolated or in small numbers and Dunn et al. 2002a for a listing of permineralized may have been produced somewhat more par- taxa, which fall into 10 genera), all of which are simoniously. built on the same basic structural plan, epito- Clearly, seeds of all sizes were a potential en- mized by Trigonocarpus (adpression) or Pach- ergy source for animals. The well reinforced 98 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133 sclerotestas of medullosans are similar to fea- environments in which large prepollen grains tures of modern seeds designed to protect them may not have had to travel far for pollination to from mechanical crushing or digestion by gut be effective (C. J. Cleal, personal communica- enzymes. And soft sarcotestas, it could be ar- tion, 2005). Water pollination is another possi- gued, served as attractants to herbivores, possi- bility (P. R. Crane, personal communication, bly vertebrates. But which ones? The evidence 2005), given the large size of the seeds and pre- for Pennsylvanian herbivory is ambiguous (Hot- pollen grains, and the semi-aquatic habitats that ton et al. 1997); animals such as Edaphosaurus, many medullosans appear to have occupied. which has large, flat, tooth batteries, have been Medullosan carbon allocation. Medullosa, suggested as seed predators. Direct evidence is sensu lato, may have been the most "expen- limited, however, and crushed seed remains have sive" plant in coal-swamp landscapes, in terms yet to be identified unequivocally in coprolites of whole-plant carbon allocation. In a study of of any kind. late Pennsylvanian plants preserved in coal balls Medullosan pollen organs, prepollen grains, from one late Pennsylvanian coal bed (and thus and pollination syndromes: The pollen organs of subject to approximately the same depositional the medullosans, highly variable in morpholog- and diagenetic history). Baker and DiMichele ical details but virtually all synangiate, have (1997) compared Medullosa, Sigillaria, and been well studied and elucidated (e.g., Millay Psaronius (but not cordaitalean woody plants) and Taylor 1979) and, in some instances debated quantitatively in terms of carbon allocation. Un- at length with regard to form and evolution (e.g., like Psaronius and Sigillaria, in which different Dolerotheca: Dufek and Stidd 1981, Roth well tissues and organs are very different in construc- and Eggert 1986, Stidd 1990). In the medullo- tion, Medullosa is uniformly rich in sclerotic tis- sans these organs were foliar borne and distinct- sue and resins, which greatly add to construc- ly individuated from pinnules. "Pollen" appears tional cost. This is in keeping with large seeds to have been, technically, prepollen; i.e., pollen and large fronds in this group, conforming with tubes were haustorial and sperm were released K-selected ecological strategies in many tree directly, presumably freely into the pollen cham- species. ber. If modern Ginkgo and Cycas serve as ac- Medullosans in the landscape. The study of curate examples of this syndrome, sperm were Pennsylvanian tropical plant community paleo- likely flagellate. Furthermore, the time elapsed ecology has a long history, given the numerous between pollination and fertilization may have exposures of plant-bearing rocks during coal been considerable, as in modern Ginkgo (Favre- mining throughout western Europe. Reviewed Duchartre 1956, Royer et al. 2003). by Scott (1977), much of the early work was Pollen/prepollen delivery systems in the pte- closely tied to biostratigraphy (Davies 1929), al- ridosperms appear, from data and inference, to though the development of quantitative ap- have been diverse. The medullosan prepollen proaches began in the mid-1900s (Dragert 1964, grains, largely attributable to Monoletes, were Pfefferkorn et al. 1975) and led to the develop- large, the longest dimension ranging from about ment of generalities that have proven robust as 100 jjLm to over 500 jjum (Taylor 1978). It seems the number of explicitly ecological analyses has impossible to imagine them being wind deliv- increased. For pteridosperms, the following gen- ered, although forms have been reported with eral patterns have emerged: vestigal sacci (Parasporites: Dennis and Eggert 1. In the early and middle Pennsylvanian (West- 1978). Recent modeling studies (Schwendemann phalian), medullosan pteridosperms are most et al. 2005) indicate rapid settling dynamics for abundant in the diversity of habitats that Monoletes, considerably different from known characterized clastic floodplain environ- forms of wind-delivered gymnosperm pollen. ments, from better drained levees and stream- Dispersal by arthropods, flying or otherwise, has sides to soggy soil areas. In such settings, been suggested, based on prepollen size alone most species appear to have been major col- (e.g., Taylor 1982). Coprolites filled with med- onists of non-swamp habitats, even if with ullosan prepollen are known (Labandeira 2000) wet soils, but only a few could tolerate areas and sloppy eating arthropods, covered with un- of long-persistent standing water. Although consumed prepollen, may have been further at- anatomical evidence to support adaptation tracted to the outer seed coats of the large med- specifically to wetlands is limited, root sys- ullosan seeds. Add to this the probability that at tems of medullosans (Rothwell and White- least some medullosans formed dense thickets. side 1974) may have aerenchyma in the small 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 99

rootlets, consistent with patterns in other include Lyginopteris hoeninghausii and plant groups from Pennsylvanian wetland Sphenopteris spp. Mariopteris-\\k& foliage systems. In addition, distributional evidence has been reported as well. The narrowness of suggests that the group as a whole was cen- this subset may reflect, in large part, the dif- tered in areas where mineral nutrients were ficulty of identifying foliage from the infor- readily available (e.g., Pfefferkorn and mation most readily available in anatomical Thomson 1982, Lamboy and Lesnikowska preservation, which may mask greater diver- 1988, DiMichele et al. 2005a, 2006). For ex- sity. ample, certain medullosans are commonly 4. In late Pennsylvanian/Stephanian age tropical abundant in organic-rich shales that probably environments of Europe and North America, formed as waterlogged mucks (e.g., when medullosan diversity appears to have de- preserved in peat substrates, medullosan re- creased or remained steady following the re- mains are commonly associated with mineral gional extirpation of the dominant middle partings in the coal bed or with elevated lev- Pennsylvanian lycopsids (Phillips et al. els of fossil charcoal: DiMichele and Phillips 1974). Some medullosans, primarily Macro- 1988). neuropteris scheuchzeri, have been recorded 2. At any local area in floodbasin settings, med- as dominant elements in clastic-rich organic ullosans rarely were notably diverse. Rather, mucks from the latest Pennsylvanian (Phillips a few species proved to be widespread at any et al. 1985, DiMichele et al., 2005a). How- given time and in any given area. Numerous ever, for the most part, they continue to play studies correlating lithofacies and plant dis- a role as subdominant species at a landscape tribution record a restricted spectrum of level. known species (e.g., Scott 1978, 1984; Gas- 5. Pteridosperms are present as vines and taldo 1985, Wnuk and Pfefferkorn 1987, ground cover throughout the Pennsylvanian DiMichele and Nelson 1989, DiMichele et al. (e.g., DiMichele and Phillips 1996b, Gastal- 1991, Demko and Gastaldo 1992, Willard et do et al. 2004). However, in the late Penn- al. 1995, Pryor and Gastaldo 2000, Gastaldo sylvanian/Stephanian, the abundance of vines et al. 2004, Falcon-Lang, in press). This may increased significantly, both in terms of spe- mean that only a few medullosan species, cies numbers and biomass. This led Kerp and based on adpressed foliage, were dominant at Krings (1998; see also Krings and Kerp any time. Most species were rare, either in 1999) to suggest that the advent of tree-fern terms of their biomass or in terms of the dominance in the late Pennsylvanian/Ste- number of sites at which they occurred in de- phanian in much of the tropics led to the clo- tectable frequencies. However, in clastic soil sure of forest canopies and attendant light at- habitats, the medullosans as a group had the tenuation compared with forests of the mid- highest recorded biomasses prior to the mid- dle Pennsylvanian/Westphalian, where open- dle of the Westphalian D, after which they canopied lycopsid forests were most remained dominant but often shared that po- common. This may have placed a premium sition with marattialean tree ferns (Pfeffer- on climbing growth forms, resulting in an in- korn and Thomson 1982). Similar patterns of crease in the abundance of climbers in late medullosan abundance and distribution ap- Pennsylvanian/Stephanian floras in general in pear to characterize coal-ball floras (Phillips Euramerican wetland habitats. et al. 1977, Phillips and DiMichele 1998). Habitat segregation patterns among medullo- 3. A subset of medullosan species, otherwise sans. Perhaps the most conspicuous pattern of known from clastic adpression preservation, habitat differentiation is that between the arbo- occurs in peat-substrate habitats where they rescent neuropterids and alethopterids. On av- are known anatomically from coal-ball pet- erage, compared with groups such as the lycop- rifactions. In other words, there are at this sids or sphenopsids, the alethopterids and the time no species that are known uniquely from neuropterids probably were much alike ecolog- peat substrates. However, the number of spe- ically, growing in high nutrient, woodland or cies reported from peats is quite small: Neu- forested environments (as opposed to standing ropteris ovata, Macroneuropteris scheu- water swamps or wet, aggradational, stream-lake chzeri, Laveineopteris rarinen'is, Alethopter- side settings). However, within the narrower is sullivantii, A. lesquereuxii, Linopteris sp. context of nutrient-rich, mineral substrate wet- Lyginopterids also known from fossil peats lands, these groups may have had different evo- 100 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133 lutionary and ecological centroids. Consider the (2002) also argued for niche segregation among study of Arens (1993), for example. She exam- genera of the neuropterids, particularly Neurop- ined the palynofloras above and below 10 fusain teris s.s. and Laveinopteris, based on species (fossil charcoal) horizons, presumably the re- abundances and patterns of turnover through mains of wildfires, in the Westphalian B Joggins geological time. They first found that species of section of Nova Scotia. Monoletes, a prepollen Neuropteris s.s. and species of Laveineopteris grain produced by medullosan pteridosperms, do not appear to have been in competition for was abundant only in sediments immediately resources. The former genus, in which they above the fire horizons. In these cases, the med- identified 20 species, apparently favored open ullosans were part of post-disturbance, coloniz- habitats where seeds could germinate in full sun, ing vegetation, presumably in open, unshaded giving it a broad range of potential habitats. Lav- habitats. Comparing the known anatomical and eineopteris, on the other hand, formed dense morphological features of Macroneuropteris thickets where its seeds would have to germinate scheuchzeri and Alethopteris spp. with extant in relatively deep shade—more a "niche spe- tropical pioneer trees, she found strong congru- cialist", in their terms, and in keeping with the ence, but also one marked difference—the large identification of hydathodes in some species of size of the seeds typical of the Pennsylvanian this genus. medullosans. Anecdotal observations in the Such findings can be extended to the species field, however, have led many authors to spec- level. Wnuk and Pfefferkorn (1984) note that ulate about the differences between Alethopteris Macroneuropteris scheuchzeri was likely a xe- and neuropterids, concluding that these lineages romorphic species (Schabilion and Reihman occupied different kinds of physical settings. 1985), consistent with Arens (1993) inferences, Wnuk and Pfefferkorn (1984), for example, not- and suggest that there were significant ecologi- ed a separation of dominance of these two gen- cal differences among neuropterids. However, era in the Westphalian D-aged Bernice Basin of xeromorphy does not necessarily mean the lack Pennsylvania. They suggested that Alethopteris of water in the environment. On the contrary, was generally the more xeromorphic of the two such plants may have grown in habitats with genera based on anatomical construction (Franks very wet substrates, presenting a different kind 1963, Reihman and Schabilion 1976). Working of physical stress; M. scheuchzeri, for example, with late Pennsylvanian age coal-ball floras, had hyadathodes (Schabilion and Reihman Mickle and Rothwell (1982) found that either 1985), suggesting excessive water. Data from Macroneuropteris-Neuropteris or Alethopteris the latest Pennsylvanian of Texas corroborate tended to be dominant in coal balls but not occur this inference. Such data indicate that M. scheu- together in equal frequency. Unpublished data chzeri, at this time and place, appears to have from the late Pennsylvanian of Texas corrobo- preferred and nearly monodominated organic rate these observations. Macroneuropteris shales that probably were mucks, formed in clas- scheuchzeri and Alethopteris zeilleri both may tic swamps with low oxygenation. Neuropteris occur in shales associated with organic accu- ovata, in comparison, at the same time and place mulations, but M. scheuchzeri is widespread, (late Pennsylvanian, north-central Texas) is rare- abundant, and mainly restricted to organic-rich ly found in organic shales, occurring in clayier beds, whereas A. zeilleri is abundant only locally lenses in association with Pecopteris, possibly and tends to occur in less organic-rich accumu- bordering sluggish channels, or in more oxidiz- lations. Co-occurrence within deposits is mini- ing channel deposits, uncommonly occurring in mal. Cuticular studies of A. zeilleri from several nearly monospecific assemblages in which N. Stephanian/late Pennsylvanian areas of Western ovata appears to have been the only tree present. Europe reveal cuticles with upper surface hairs Neurodontopteris auriculata is also an element and stomata restricted to the lower surfaces of of this same species pool, but occurs yet more pinnules. On the basis of such cuticular data, rarely, its association extending into assemblag- Kerp and Barthel (1993) interpret this plant as es that include conifers and Sphenopteridium mesophytic, growing on the margin of organic (sensu Mamay 1992, a probable seed plant char- accumulating areas. Thus, although these two acteristic of moisture-stressed habitats of the species may have overlapped within the overall Mississippian, rediscovered by Mamay in late species pool, there appears to be an average dif- Pennsylvanian and Early Permian seasonally dry ference in their ecological centroids. floodplain floras) both xeromorphic groups, sug- Cleal and Shute (1995) and Shute and Cleal gesting that A^. auriculata was the medullosan 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 101 most tolerant of moisture stress, at least in these be speculated that changes in regional patterns landscapes. of moisture distribution, particularly drying, At much larger scales of analysis, patterns of could have affected the ability of populations to medullosan diversity and species turnover can maintain both critical size and spatial continuity, reveal broad trends in ecological tolerance. Cleal which may be the key to many of the species and Shute (1995) is an exemplar of such a study. extinctions. Species that survive and transgress After extensive reassessment of neuropterid tax- times of major environmental change, L. rari- onomy and stratigraphic distribution, they were nervis and M. scheuchzeri for example, are able to use the refined systematic database to strongly associated with organic accumulating, examine diversity patterns and possible paleo- swampy conditions, and occur both in wet clas- ecological controls on diversity. Looking mainly tic substrate and wet peat substrate habitats. at species from Western Europe, Cleal and Shute These broad, catholic distributions in wetlands found that neuropterids show marked correlated may have buffered them against the regional changes in diversity corresponding to inferred wetting and drying trends in the later middle broad-scale climatic changes in the Pennsylva- Pennsylvanian. Similar patterns are found in the nian tropical lowlands (Phillips and Peppers survival patterns of tree ferns and other wetland 1984, Gastaldo et al. 1996, Cleal et al. 1999). plants from the Carboniferous into the Permian, Species of Neuropteris s.s., in particular, were where the "Carboniferous" taxa survive well restricted to the wetter intervals. With the initial into the Early Permian in "wetspots" within the onset of long-term drier tropical conditions, neu- drier landscape (Kerp 1996, 2000; DiMichele et ropterids with flexuous veins begin to appear, al., 2006). ultimately leading phylogenetically to forms Inferring medullosan environments of growth. with reticulate venation (Reticulopteris). Simi- Although primarily plants of wetlands, medul- larly, in Laveineopteris there is a change in the losans also appear to have grown in habitats predominant species at the points of climatic with some seasonal water limitation. Linking transition, although the most common late mid- morphology to physiology is often ambiguous, dle Pennsylvanian species, L. rarinen>is, shows however, and the determination of ancient hab- no response to such inferred changes in climate. itats of growth should be made with caution Macroneuropteris, a genus with only four re- from morphology alone. For example, reticulate ported species, only two of which are well venation has been suggested to enhance physi- known, also seems robust to macroclimatic ological functioning under water-limited condi- changes, possibly because the entire species tions, and appears in parallel in several groups group had similar ecological tolerances. The ge- of medullosans. Yet, sedimentology and other nus Paripteris, which Laveine (1997) assigns to aspects of anatomy conflict with this generaliza- the Perispermaceae, also undergoes a marked tion for some species. For example, the reticu- change in species representation at the initial on- late-veined Reticulopteris muensteri also has hy- set of widespread climatic changes at the early- dathode-like structures (Reihman and Schabilion middle Pennsylvanian transition; the newly ap- 1978), anatomical modifications of leaf vein pearing species go extinct with the presumed re- endings that permit water to be exuded without turn of wetter conditions in the later part of the transpiration. Hydathodes occur in plants that middle Pennsylvanian, giving rise to a reticulate grow under conditions of high soil moisture and veined form, Linopteris (the foliage of the stem low transpiration and, thus, help to keep water genus, Sutcltffia), which is common through the moving through the plant. Zodrow and Cleal climatic changes of the middle-late Pennsylva- (1993) argued that the reticulate venation of R. nian boundary and into the late Pennsylvanian muensteri places it ecologically in slightly drier (Stephanian). Finally, the species of Neuraleth- parts of floodplains. However, Willard et al. opteris appear to have been strongly confined to (1995) identified pinnules with venation similar wetter conditions and disappeared with the onset to R. muensteri in dark organic shales immedi- of climatic drying. ately above the Springfield Coal bed, Westphal- In light of these patterns, Cleal and Shute ian D of Indiana, interpreted as standing-water (1995) also noted that "climate" per se may not clastic swamps, and suggesting growth of this be the sole arbiter of such diversity trends. Cor- species in water saturated soils. Phillips and rectly, it seems to us, they note that climate will DiMichele (1998) identified (incorrectly as Re- interact with topographic factors, exacerbating ticulopteris) Linopteris in a thin coal, also above the effects of landscape heterogeneity. It might the Springfield Coal bed, in association with the 102 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133 uncommon petrified medullosan stem Sutcliffia, Growth habits and ecology of the Pennsyl- raising the possibility that reticulate veins are vanian Lyginopteridaceae. Early Wesphalian not necessarily a sign of serious water limitation coal-ball specimens of Lyginopteris helped pro- in these plants either. On the other hand, the re- vide the concept of the "seed fern." Oliver and ticulate-veined Barthleopteris germarii lacks hy- Scott (1904), using permineralized specimens, dathodes and has cuticular morphology (papil- demonstrated the common taxonomic affnities late stomata and glandular trichomes) that led of the seed Lagenostoma lomaxii, the stem Ly- Krings and Kerp (1998) to place it in periodi- ginodendron, and foliage of this plant, which, in cally water-limited environments. adpression, had been attributed to the forked and As an aside, reticulate veins also have been a highly divided, "fern-like" fronds Lyginopteris source of taxonomic confusion, only recently (Sphenopteris) hoeninghausii. Other genera re- clarified by cuticular analysis. Reticulopteris ferred to this family include the following plants muensteri, for example, is the only species of all largely circumscribed on the basis of anatom- the genus and, through cuticular analysis, shows ical preservation: Rhetinangium arberi, Heter- clear affinities with Neuropteris of the A^. gen- angium spp., Schopfiastrum decussatum, and uina lineage (Zodrow and Cleal 1993). Barthel- Microspermopteris aphylla. All are believed to opteris germarii, originally described as a spe- bear small fronds (up to 1 m in length, but per- cies of Reticulopteris, has cuticlar morphology haps £ 30 cm in some species: Gastaldo 1988) that places it closest to Neurocallipteris (Zodrow with finely dissected foliage of the Eusphenop- and Cleal 1993). teris or Lyginopteris type, bear small seeds (£ Hydathodes have been described in a number 5 mm), have various configurations of scleren- of different pteridosperms, including Laveineop- chymatous plates of small diameter in the cor- tical tissues, and monostelic stems. These plants teris rarinen'is (Oestry-Stidd 1979), Macroneu- are frequently preserved as charcoal in coal-ball ropteris scheuchzeri (Schabilion and Reihman permineralizations, far more frequently than 1985), Lescuropteris genuina (Krings and Kerp would be expected given their generally small 1997), Mariopteris occidentalis (Krings et al. biomass in coal-ball assemblages. 2001), as well as Reticulopteris muensteri. Lyginopteris, the basis of the Lyginopterida- These structures do not appear to be unambig- ceae, is considered to have been a facultative uous indicators of high soil moisture, however, climber (Fig. 1), based on a wide range of evi- and need to be interpreted in light of plant habit dence. The plant may have been one of the more and likely conditions of growth. Lescuropteris robust Carboniferous plants with such habit, its genuina and M. occidentalis, for example, are stem commonly reaching 3.0—4.5 cm in diame- interpreted as climbers, plants that generally ter but bearing relatively small fronds. Biome- grow under conditions of high insolation and chanical analysis suggests that the stem was not thus water stress. Hydathodes in M. scheuchzeri self-supporting (Speck 1994), which is consis- are consistent with its growth in swampy set- tent with the generally low abundance but wide tings, especially in the later Pennsylvanian occurrence of the plant. Its habit was probably where it has been identified as a dominant ele- shrub-like to scrambling where trees were lack- ment in organic shales and coals of the latest ing, perhaps forming thickets (Gastaldo et al. Pennsylvanian in Texas (Phillips et al. 1985, 2004). Because of the distinctive anatomy and DiMichele, unpublished data). Yet, Schabilion morphology of the stems and fronds, Lyginop- and Reihman (1985) argued, in apparent contra- teris fossils are identifiable in both adpression diction with their recognition of hydathodes in and permineralized modes. In the classic study the species, that M. scheuchzeri had xeromorph- of Oliver and Scott (1904), Lyginopteris old- ic leaf anatomy. Macroneuropteris scheuchzeri hamia was correlated with Lyginopteris (Sphen- has a long stratigraphic history and may have opteris) hoeninghausii partially on the basis of varied considerably in its habitat preferences distinctive capitate secretory glands on the stem, during that time. There do appear to have been frond, and seed cupule. These glands both aided changes in size, in the density of leaf hairs, di- in recognition of the whole plant assemblage minishing over time, and in the shape of the pin- and raised questions about the gland functions nule base. Perhaps this "species" is, in fact, a as animal deterrents, attractants, or satiation sub- complex of closely related species not easily dif- stitutes (Phillips 1981, Krings et al. 2003a). ferentiable from leaf morphology alone (Darrah Lyginopteris was an important plant in both 1969). peat-forming habitats and in clastic wetland en- 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 103 vironments in the late Mississippian and early near the Westphalian A-B boundary suggests Pennsylvanian of Europe, where it is most com- some kind of complex ecological cause, possibly mon in clastic environments associated with ma- the conjunction of several things, for demise of rine bands (e.g., Evans et al. 2003). It was long such a cosmopolitan genus. thought not to be present in American floras. Schopfiastrum decussatum (Rothwell and However, it was reported and figured in com- Taylor 1972, Stidd and Phillips 1973) is very pressions by Gillespie and Pfefferkorn (1979) likely a climber. It consists of small diameter (3 and Blake et al. (2002), and anatomical studies cm or less), woody stems that bear relatively have found it and closely related forms to be large petioles, inferred from their diameters in present as early as the Upper Mississippian/low- permineralized specimens, with some similari- er Namurian A, in deposits from Arkansas (To- ties to Mariopteris according to Stidd and Phil- mescu et al. 2001, Dunn et al. 2003a). It also lips (1973). The fronds are borne alternately and occurs in coal-ball deposits from a Westphalian distichously. The stem cortex contains radially A (Langsettian) coal of Alabama (Winston and elongated, longitudinal sclerenchyma bundles, Phillips 1991). Gastaldo et al. (2004) found Ly- giving the stem strength but flexibility, typical ginopteris hoeninghausii to be very abundant to of plants with climbing habit. Schopfiastrum de- dominant locally in an in situ, spatially pre- cussatum is known from the middle Pennsylva- served, drowned, swamp forest of early Penn- nian of North America. The plant is almost iden- sylvanian age in Alabama. As with others, these tical anatomically and in general foliage mor- authors inferred Lyginopteris to be a climber or phology to Rhetinangium arberi, which was de- ground cover, probably a thicket-forming plant, scribed from the Namurian of Britain (Gordon widespread but most abundant in areas where 1912). However, the latter plant bears its fronds the sparseness of trees suggested high light helically rather than in a decussate manner. availability, paralleling the abundance of vines Microspermopteris is a small lyginopterid and ground cover in modern forests. pteridosperm that spans most of the Westphalian Lyginopteris is a more common and abundant (lower and middle Pennsylvanian). It is known component of Westphalian A/Langsettian ad- exclusively from coal balls (Pigg et al. 1986). pression floras, representing the wide array of Anatomically, the plant is most similar to Het- floodbasin habitats, than it is in peat-forming en- erangium. However, the maximum known stem vironments, represented by coal balls. The coal- diameters are only about 1 cm and stems have ball distribution of Lyginopteris oldhamia is proven to be highly sinuous. The frond of the quite variable among the coal-ball collections plant appears to bear foliage most similar in from England and Belgium, being most abun- form to the genus Sphenopteris, but with a uni- dant at the famous Shore locality in the Lanca- form internal histology, lacking differentiation shire coal field but much less conspicuous at into palisade and spongy mesophyll. Scattered other heavily collected sites in the same coal bed resinous cells may be present within the lami- (Phillips et al. 1985). In the American coal-ball nate portions of the foliage. Fronds do not fork, deposit from Alabama (Winston and Phillips which is unusual in the lyginopterids. Overall, 1991) Lyginopteris is common but not particu- the growth habit appears to be that of a gracile larly abundant, accounting for about 1.5% of the vine or scrambling ground cover. Pigg et al. total peat biomass. The plant may be most abun- (1986) suggest that the small seeds Conostoma dant in ecotonal areas where nutrient levels were villosum, which have conspicuous integumenta- higher than in subenvironments with long peri- ry flanges and trichomes, similar to such fea- ods of standing water. For example, Phillips tures found on Microspermopteris stems, may (1981) noted the highest abundance of Lyginop- belong to these stems. Lack of attached repro- teris (9% of biomass) in the Katharina seam of ductive organs, or regular association with any the Ruhr, where the lycopsid Paralycopodites particular seeds or pollen organs, suggests infre- brevifolius was abundant. This was the highest quent reproduction, possibly confined to spe- reported stratigraphic occurrence of Lyginopteris cialized environmental conditions. The plant in coal balls. may have mainly spread by vegetative propa- Patteisky (1957) developed a detailed bio- gation, typical of a vine or scrambler. stratigraphic zonation for Lyginopteris from Heterangium is another of the small, probable goniatite zones from the Visean up into the climbing or scrambling lyginopterids known ex- Westphalian B in Western Europe. The sudden clusively from anatomical, coal-ball preserva- extinction of Lyginopteris across all Euramerica tion. Numerous species have been described 104 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

(Pigg et aL 1987), the oldest from the mid-Mis- It is interpreted as a climber (Krings et al. 2001) sissippian. Plants are often found preserved as based on small hooks on the lower/abaxial sides fossil charcoal, suggesting that these species, of the pinnae, a pattern that differs from all other much like other small climbers or ground cover, known mariopteroids, which have climber hooks occupied environments frequently swept by on apical extensions of the pinnae. The exten- fires. Sphenopteris obtusiloba has been de- sive distribution of these hooks probably means scribed as the foliage of Heterangium american- the plant climbed in dense vegetation or on del- um (Shadle and Stidd 1975) and Rhodeopteri- icate foliage, given the large number of small, dium (as Rhodea) has been reported from a Mis- widely dispersed hooks. In Oklahoma, M. occi- sissippian species of Heterangium (Jennings dentalis tends to occur alone in abundance on 1976). rock slabs and is not so common in mixed as- semblages (DiMichele et al. 1991). It is possible The Mariopteridaceae: climbing and thicket- that it formed thickets. Vein endings are thick- forming habit. The Mariopteridaceae is a puta- ened and may have functioned as hydathodes to tive family of pteridosperms all of which appear help facilitate water flow in young fronds. These to have had climbing habits (Krings et al. are very common in scrambling/climbing pteri- 2003b). Plants of this group are known from ad- dosperms. This species also has low stomatal pression preservation and have not been well density, small stomatal apertures, and sunken documented anatomically, although Mariopteris guard cells, which also indicate water stress, has been reported as the foliage of the vine-like typical of a vine (Krings et al. 2001). The wide- Schopfiastrum (Stidd and Phillips 1973), consis- spread, middle Pennsylvanian species Mariop- tent with the inferred habits of these plants. In- teris muricata was shown by Barthel (1962) to cluded are the genera Mariopteris, Karinopteris, have numerous hairs on the upper surface of its Pseudomariopteris, and Helenopteris. The leaf laminae, which could be considered to be strong morphological similarities among these in keeping with a xeromorphic, climbing or plants indicate close phylogenetic relationships thicket-forming habit for members of this clade. and their climbing habit can be seen as an ex- There are two species in the genus Pseudo- ample of phylogenetic/historical constraints on mariopteris, both of which appear to have been ecology. The species that have been identified as climbers or thicket formers, based on the pres- climbers use small hooks on the ends of lateral ence of climber hooks (Fig. 6). Pseudomariop- pinnae or the undersides of fronds to attach teris busquetii is known from the late Stephanian themselves to each other or to other plants, cre- of France, the upper Rotliegend of Germany, ating thickets or climbing on the organs of other and the latest Pennsylvanian/earliest Permian of plants. Most of the species exhibiting these fea- north-central Texas, a restricted stratigraphic tures are geographically and/or stratigraphically range overall. Krings et al. (2001) determined restricted in their distributions. One species, that the plant had a climbing habit based on Pseudomariopteris busquetii, has been found to stems 1.0—1.5 cm in diameter bearing forked bear small, platyspermic seeds on the underside fronds 15—30 cm in length. Such narrow stems of fronds, which is similar to the disposition of and relatively large fronds suggest that the seeds in the scrambling plant Dicksonites (Cal- plants were not self-supporting. Support came listophyton). This suggests a relationship of the from climber hooks borne on apical extensions Mariopteriaceae and the Callistophytaceae, in of pinna axes. In addition to, or perhaps more the Callistophytales (Krings et al. 2001), assum- likely than, climbing, plants of this species, as ing that the mariopterids actually are related as with most of the mariopterids, appear to have a monophyletic group. formed dense thickets in which the plants were Mariopteris is the largest genus of the family. mutually supporting. This interpretation is fur- Species in the genus initially were identified as ther borne out by observations of patchy dense having simple "spine-like prolongations" of the accumulations of Pseudomariopteris busquetii lateral pinnae (Boersma 1972), which, under in late Stephanian equivalent deposits from closer scrutiny, have proven to bear small spines north-central Texas. In these deposits, P. bus- or hooks that facilitate attachment to other quetii is far and away the dominant species, rais- plants. For example, M. occidentalis, which is ing questions of how such large, local biomass known mainly from the eastern and midwestern could be derived from climbing plants (climbing United States, reached its zenith of abundance on what?). Pseudomariopteris cordato-ovata has during the middle Pennsylvanian in Oklahoma. small fronds with paired climber hooks at the 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 105

on club-shaped tips of larger hooks. DiMichele *-1 et al. (1984) described a Karinopteris species t from the Indiana "paper" coal that had small bifurcate fronds, perhaps as much as 25 cm in w length, with robust hooks borne on the abaxial sides of pinna prolongations. Kerp and Barthel (1993) reexamined cuticles of this plant and de- termined that it was highly xeromorphic. Upper leaf surfaces were found to be hairy with thick cuticle and lacking in stomata, which were re- stricted to thinly cutinized lower leaf surfaces where they were likely sunken and overarched by papillae. The remains of these plants occur in dense accumulations in restricted deposits— the plant locally dominates the flora of the In- diana paper coal, which, again, might indicate formation of dense thickets in which the plants were hooked onto one another for support. The undescribed species appears to be both strati- graphically and geographically restricted. A report of climber hooks in a plant described as Eremopteris lincolniana by White (1943; dis- cussed in Krings et al. 2003b—likely not an Er- emopteris according to C. J. Cleal, personal communication) suggests that this mode of climbing evolved in other lineages. The possi- bility of mariopteroid affinity for this plant also FIG. 6. Pseudomariopteris busquetii, a climbing or tliicket forming plant, reconstructed here as climbing should be considered, given the unusual nature a meduUosan pteridosperm with free-standing growth of the climber-hook morphology and its strong habit. Reprinted from Krings et al. (2001, Fig. 12) with expression among this phylogenetically related permission from the Botanical Society of America. group of plants (Gastaldo and Boersma 1983a, b). Krings et al. (2003b) noted that this kind of end of spine-like prolongations of pinnae, and, thus, also appears to be a climber or thicket- climbing or thicket-forming habit, one facilitated former, like P. busquetii (Krings and Kerp by the presence of small numbers of small 2000). hooks, is most common today among angio- Krings and Kerp (2000) segregated the genus sperm vines and lianas of medium to small stat- Helenopteris from Pseudomariopteris. Athough ure that grow in areas of dense vegetational cov- the single species, H. paleaui has not been found er. A parallel ecomorphological syndrome was to have climber hooks, it has been interpreted as described for several adpression species of the a climber/thicket-former on the basis of small small sphenopsid Sphenophyllum by Batenburg stems bearing fronds with a maximum length of (1982), which bore hooks at the end of leaves 20 cm, and on its clear phylogenetic affinity to and appear to have formed dense thickets. In other members of the mariopterid clade. The either instance, pteridosperm or sphenophyll, the species has a restricted geographic distribution, possibility has been raised of much denser found for certain only in the late Stephanian of ground cover on Pennsylvanian-age landscapes, central France. particularly those of flood-basins with wet min- Karinopteris follows the morphological and eral soils, than is traditionally shown in recon- distributional patterns found in other members structions. Large areas of such landscapes may of the mariopterid group. Schultka (1995) re- have been occupied by dense monospecific ported Karinopteris acuta from the Namurian B stands of plants with facultative thicket-forming of Germany, which is a very early occurrence to climbing growth habits—Paleozoic kudzu, of mariopterid climbing plants. The species ap- paralleling the recent explosion of that plant pears to be a climber with many booklets borne throughout much of the southern United States 106 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

plant sperm delivery (Rothwell 1972). The over- all aspect of this plant is one of an opportunist: wind pollination followed by the production of many, highly dispersible, small seeds, patchy, often dense occurrence within wetland land- scapes, and frequent preservation as charcoal, as if growing in areas prone to fires. Species of the genus appear focused on the colonization of available physical space and thus the garnering of light and mineral resources by space expro- priation.

Johnhallia, a possible relative of Callistoph- yton. Johnhallia lacunosa is a small pterido- sperm (Stidd and Phillips 1982) similar in ana- tomical form to Callistophyton but differing suf- ficiently that the authors considered its relation- FIG. 7. Callistophyton sp., a scrambling ground- ships ambiguous and possibly distant. Stems are cover plant. Reprinted from Rothwell (1981, Fig. 1) small (about 1 cm in diameter) and bear small with permission from Elsevier. fronds—the petiole bases are 2.5 mm in diam- eter. Foliage is delicate without internal cellular division into palisade and spongy mesophyll; it after its introduction from Asia for soil erosion appears to be of the RhodeopteridiumlDiploth- control. memalPalmatopteris type, which is nearly non- The ecology of Callistophyton. Callistophy- laminate in form. Growth habit is not known. ton is the anatomically preserved equivalent of Johnhallia lacunosa is quite rare, although it the adpressed genus Dicksonites. Thanks to de- may have been confused with Callistophyton tailed anatomical studies, organs of the plant and so under-reported. As with many of the have been correlated and one of the best "whole small pteridosperms known from anatomical plant" concepts of any Pennsylvanian plant has preservation in coal balls, specimens are often been developed (Rothwell 1981). Recently, Gal- preserved as charcoal. tier and Bethoux (2002) have further refined un- derstanding of the growth habit. There are two Peltasperms in Permian landscapes. The anatomically circumscribed species of Callisto- peltasperms are a group of pteridospermous phyton, C boysettii, from the middle Pennsyl- plants that were thought to be largely Mesozoic vanian and latest Pennsylvanian-earliest Permian in distribution until Kerp (1982) determined that of North America and Europe, and C poroxy- the widespread and well known callipterids were loides from the late Pennsylvanian of North associated with peltaspermous-type reproductive America. In both instances, the plants have a organs. Since that time, it has become increas- scrambling to climbing habit and could form ingly appreciated that this was a diverse group dense stands (Galtier and Bethoux 2002) (Fig. with complex morphologies, widely distributed 1, 7). They bore bifurcate fronds to which the primarily in the Early Permian tropics and in the seeds, known as Callospermarion in petrifac- latest Early through Late Permian in the tem- tion, were attached abaxially to pinnules (Lan- perate regions (Naugolnykh and Kerp 1996, giaux 1986). Seeds were on the smaller end of Naugolnykh 1999). These plants are known ex- the pteridosperm range, 4—5 mm in length, clusively from foliar and reproductive adpres- which is still rather large when compared with sions. It can be supposed from the shapes, sizes, many modern flowering plants. The seeds were and epidermal features of the known organs that borne in great abundance (Grand'Eury 1905). the plants assumed a variety of growth forms. Bisaccate pollen, attributed to Vesicaspora, was At present, however, the growth habits of these produced in foliar borne synangia. It was deliv- plants are effectively unknown. ered by wind to the ovules and captured by a The peltasperms consist of several major plant pollination droplet. The male gametes were de- groups: the callipterids (probably many unde- livered to the eggs via a pollen tube, making this scribed forms, the best known being Autunia and the earliest known occurrence of modern seed- Rhachiphyllum), the comioids (Comia and pos- 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 107 sibly one or more undescribed genera), the su- 1996). On the other hand, Rhachiphyllum schen- paioids (Supaia, Glenopteris, Compsopteris, kii (Kerp 1988) occurs in association with Cal- Protoblechnum, Brongniartites), and probably amites gigas on floodplains (Barthel 1976, Bar- (and perhaps most controversially) the 'Ameri- thel and Roessler 1994, 1996). Other species ap- can' gigantopterids (as opposed to the forms de- pear to have been more strongly confined to bet- scribed from the Cathyasian areas, sometimes ter drained substrates. under the same generic names) (Gigantopteri- Unfortunately, it has not been possible to dium, "Cathaysiopteris", "Zeilleropteris", Ev- make clear distinctions among most of the major olsonia, Delnortea, and Lonesomia). The tax- peltasperm lineages as regards particular envi- onomy of these plants has been summarized in ronmental tolerances or ecological centroids, papers by several authors (Kerp 1988, Kerp and which would be related to growth habits as well Haubold 1988, Naugolnykh 1999, Kerp et al. as intrinsic physiological differences. The callip- 2001, Krings et al. 2005, DiMichele et al. 2006). terids are the most widespread of this group of Members of all these plant groups are abun- plants. The full range of species diversity in this dant in the American southwest (Arizona, Kan- group has yet to be fully understood and the pat- sas, New Mexico, Oklahoma, Texas), which tern of broad geographic and environmental oc- may serve as a model for the basic aspects of currence seems to reflect several species and their paleoecology; similar patterns have been genera (Kerp and Haubold 1988) with slightly identified in Europe (Kerp and Fichter 1985). In different environmental tolerances. The Ameri- these environments, the fossils can be closely can gigantopterids, on the other hand, are rela- associated with depositional environments (Read tively rare, occurring generally in low abun- and Mamay 1964, Nelson et al. 2001, Mack et dance, even if widely distributed. Furthermore, al. 2002, 2003) and with associated indicators of the dominant genera of this clade have only lim- paleoclimate, such as paleosols and geochemical ited spatio-temporal overlap, suggesting possible estimates of temperature (White 1929, Mack et environmental differences or replacements in al. 1991, Tabor and Montanez 2004, 2005). In- time-space through background extinction and dicators of depositional environments place the evolution. The comioids illustrate a pattern sim- plants mostly in streamside environments, with ilar to the gigantopterids—there are at least three preservation occurring in small lakes, crevasse forms in the southwestern U.S.A. that are com- splays, or slack-water deposits in bars of active- plementary in space and time. ly flowing channels. Associated indicators of lo- Some of the more peculiar patterns of occur- cal climate consistently point to seasonality, of- rence are to be found among the supaioids. For ten with the possibility of severe drought at example, in the Abo Formation of New Mexico, times, indicated by calcic vertic paleosols, cali- deposits have been found that are uniformly che paleosols, molds of evaporate crystals in monotypically dominated by Supaia thinnfel- plant-bearing mudstones, and associated evi- dioides through several meters of crevasse-splay dence of strongly seasonal flow and surface ex- sandstones, suggesting large areas of seasonally posure of deposits in plant-bearing channels and flooded forests occupied nearly entirely by one crevasse splays. Recent studies on paleotemper- species of peltasperm. Yet, in other parts of the ature patterns in the Early Permian of Texas and southwest, such deposits have not been seen, Oklahoma (Tabor and Montaiiez 2005), suggest and Supaia, in general, does seem to be concen- that significant increases, of at least 10° C ac- trated in its ecological dominance to the farthest companied a shift from dominance of basinal western parts of the Early Permian tropical re- wetlands by "Pennsylvanian" floras, rich in gion. The distinct spatial distribution in North medullosan pteridosperms and marattialean tree America of such plants as Supaia and Glenop- ferns, to "Permian" floras, rich in peltasperms teris, proposed by Read and Mamay (1964), has and conifers. not proven to be as robust as originally pro- Detailed studies of European callipterids have posed, particularly for Supaia, which more re- found distinct environmental preferences for cently has been reported from the Permian in some species (Kerp 1988). Autunia conferta, for widely separated parts of the world (e.g., Wang example, is commonly found in environments 1997). However, when viewed globally, there that are drier or better drained and only excep- are centers of distribution of these plants, at least tionally in swampy deposits, such as dark or- as their taxonomy is presently understood. Pro- ganic shales. Morphological features of this spe- toblechnum, for example, is best known from cies support this environmental inference (Kerp China (Halle 1927, Sun et al. 1999), whereas 108 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

Naugolnykh (1999) has argued that Compsop- population size, developmental instabilities, ran- teris is restricted to the Angaran floral realm, dom walk to extinction in small founder popu- and Glenopteris (Sellards 1900) is largely con- lations, etc., the potential for survival may be fined to Kansas, with some questionable occur- orders of magnitude higher, even if still quite rences in surrounding states (Krings et al. 2005). low, in such "empty" habitats. Once a radiation These genera are quite similar in many aspects takes place on some body-plan theme, resource of their morphology and taxonomic refinements saturation is expected to increase and the poten- are needed to work out the biogeographic pat- tial for the survival of innovations will decrease terns. Such refinements may reveal that apparent significantly (Valentine 1980). regional differences are, in fact, a consequence Early land plants appear to have accrued mor- of poorly understood taxonomy. phological complexity only gradually, radiating In summary, the peltasperms appear to have extensively once some critical level of complex- escaped restriction to wetland environments and ity and associated developmental control was at- to have undergone a significant morphological tained (Knoll et al. 1984, DiMichele and Bate- radiation in seasonally dry environments during man 1996). Once major morphological changes the Early Permian. The initial phases of this ra- were possible, however, plants began a morpho- diation appear to have begun in the late Penn- logical radiation in the Middle Devonian that led sylvanian, based on reports of callipterids from to the emplacement of all major body plans by that time (Remy 1975 and Remy et al. 1980, the end of the Devonian or perhaps into the ear- reported "Callipteris" flabellifera from the liest Mississippian (DiMichele et al. 2001). Dur- Stanton Limestone of Kansas, which is consid- ing this radiation, major clades partitioned the ered to be late Pennsylvanian in age on the basis terrestrial habitat, with the seed plants and the of marine fossils), interestingly in association ferns becoming most prominent in terra firma with primitive conifers. The tolerance of season- settings. Of the Late Devonian through Permian al drought may have been a key factor in per- seed plants, as demonstrated by the phylogenetic mitting elements of this group to move into, and analysis of Hilton and Bateman (this volume), become ecologically significant components of, nearly all can be considered part of the paraphy- higher latitude, temperate environments subject letic pteridosperm ("seed fern") clade, espe- to light and temperature seasonality. Many of cially during the Mississippian. the important genera were first described from The signature pattern for evolution of new high-latitude Late Permian assemblages and pteridosperm families and genera is the appear- only later were discovered to be abundant in ance of new forms in disturbed and extrabasinal Early Permian tropical habitats. areas. During the early Mississippian (Tournai- sian) for example, pteridosperms are both abun- Pteridosperms, Ecology and Evolution. dant and diverse elements in highly disturbed The evolution of innovation in pteridosperms landscapes where preservation was made possi- follows a pattern that has become increasingly ble, perhaps, only because of associated volca- apparent in studies of the relationship between nism, giving us a look at vegetation types we Paleozoic terrestrial plants and what, for lack of might not ordinarily see (e.g., Scott et al. 1984, a better descriptor, might be termed "landscape Galtier et al. 1988, Bateman and Scott 1990, position." It appears that speciation, particularly Bateman 1991). Early ferns follow a similar pat- of a magnitude that leads to major new body tern, appearing in highly disturbed environments plans, what has been termed "saltational" evo- (Scott and Galtier 1985), thus settings with con- lution (Bateman and DiMichele 1994b), is most stant areas of resources available to species with likely to occur in environments that are resource opportunistic life histories. Equally interesting is undersaturated. In such settings, there are op- the appearance of larger, somewhat more "K- portunities for survival of radically new forms selected" tree pteridosperms at this same time, because of reduced incumbent advantage (Val- appearing as isolated pieces of wood potentially entine 1980, Gilinsky and Bambach 1987)—if transported from more remote, extrabasinal set- plants are already present in such settings, it is tings (Galtier 1992). Evidence of deeply rooted probable that resources remain for exploitation paleosols as early as the Late Devonian (Algeo by new innovations. Even though it might be et al. 2001) also suggests that some linages were expected that survival of novel morphologies evolving the potential to escape from the heavily will be low in any kind of environment, due to populated wet lowland habitats, so abundantly anything from exceedingly slow increases in preserved in the terrestrial fossil record. 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 109

By the Pennsylvanian, the record of pterido- (Broutin et al. 1990, DiMichele and Aronson sperms was mostly that of lowland wetlands. We 1992). can only speculate about the origin of the di- With the advent of seasonal climate regimes verse groups of pteridosperms that populated in the western tropical belt during the latest these environments because extrabasinal floras Pennsylvanian and Early Permian, the flora un- are rare. It is possible that many, such as the derwent dramatic changes. However, the plants medullosans, originated in what were initially that dominated the newly appearing landscapes resource undersaturated habitats during the late did not occur as gradual invaders, displacing the Mississippian, or by exploitation of life roles Pennsylvanian elements from increasingly re- that permitted resources to be exploited in novel stricted wet sites. Rather, the change was one of ways, such as through the evolution of climbing wholesale biome replacement—plants track cli- habit, which appears to be ancestral in this group mate in groups, reflecting the discontinuities in (Dunn et al. 2003b). Subsequent evolution re- climatic conditions (Ziegler 1990). As a result, flects the carving up of portions of the lowland a new flora became emplaced in the depositional wetlands, which were the environments in which basins of the tropics, one rich in such pterido- the model pteridosperms, the medullosans and sperm groups as peltasperms, which harbored lyginopterids, reached their zenith. enormous diversity. The peltasperms are among There is, however, evidence that evolution many lineages thought to have been exclusively continued in seasonally dry, physically stressful Mesozoic in distribution, but which have since marginal areas and that such areas also may been discovered to have Permian ancestry (Kerp have harbored elements of much older pterido- 1988). This pattern, now found for a number of sperm lineages, serving as refugia, as the tropi- lineages, suggests that considerable body-plan cal areas became increasingly suitable for plants level macroevolution was taking place in re- tolerant of moist climates. Consider, for exam- mote, extrabasinal areas (Broutin et al. 1986, ple, the genus Sphenopteridium, which has been Kerp 1996, 2000; DiMichele et al. 2001), where attributed to some of the larger early arborescent resource underutilization may well have been pteridosperms. Characteristic of the Mississip- the key to evolutionary opportunity. It is the taxa pian, it occurs then in plant associations that ev- that evolved in these remote areas, and that ap- idence seasonal climates and moisture deficits. peared "fully formed" when climatic conditions This genus appears to jump over most of the changed and allowed them to occupy the low- fossil record of the Pennsylvanian tropics, to re- lands, that persisted into the Mesozoic to be- appear late in the Pennsylvanian in isolated as- come significant elements of those floras. semblages, and continue into the Early Permian Although the pteridosperms are good indica- as a component of floras rich in conifers and tors of this basic spatial evolutionary model, it peltasperms, typical of warm, seasonally mois- does appear to be generalizable to seed plants, ture-limited environments (Mamay 1992). and probably ferns. The patterns become more The transition between the Pennsylvanian and muddled in the later Mesozoic as the diversity Permian, in the tropical realm, is an extended of land floras explodes with the advent of the one. The intercalation of floras typical of ever to angiosperms. Thus, these ancient lineages, in the mostly wet climates with those of seasonally dry less diverse world of the Paleozoic, may be climates begins in a small way as far back as among the clearest indicators of some of the ba- the late Middle Pennsylvanian, and is detectable sic relationships between plant evolution and the through such robust plants as conifers, whose physical environment. foliage could survive transport and thus give ev- idence of their presence in the hinterlands long Literature Cited before they appeared in the lowlands as part of autochthonous or parauthochthonous assemblag- ALGEO, T. J. AND S. E. SCHECKLER. 1998. Terrestrial- marine teleconnections in the Devonian: links be- es (Lyons and Darrah 1989). Although floras tween the evolution of land plants, weathering pro- typical of on-average wetter and drier climatic cesses, and marine anoxic events. Phil. Trans. Roy. regimes are intercalated during the late Penn- Soc. London B 353: 113-130. sylvanian, and even into the earliest Permian, ALGEO, T. J., S. E. SCHECKLER, AND J. B. MAYNARD. 2001. Effects of Middle to Late Devonian spread they exchange few species, and then primarily of vascular land plants on weathering regimes, ma- those of streamside wet habitats, such as sphe- rine biotas, and global climate, p. 213—236. In P. nopsids and opportunistic/weedy tree ferns G. Gensel and D. Edwards [eds.], Plants Invade the 110 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

Land: Evolutionary and Environmental Perspec- fructifications with particular reference to the Med- tives. Columbia University Press, New York, NY. uUosae. Ann. Missouri Bot. Gard. 36: 287—352. ALLEMAN, V., H. W. PFEFFERKORN, AND D. M. ERWIN. BEEHLER, H. E. 1983. Anatomy and frond architecture 1995. Los contextos paleoecologicos variados de of Neuropteris ovata and Neuropteris scheuchzeri las floras Carbonlferas de Paracas (lea, Peru). Bol- from the Upper Pennsylvanian of the Appalachian etin de la Sociedad Geologica del Pent 84: 37—42. Basin. Can. J. Bot. 61: 2352-2368. ANDREWS, H. N. 1947. Ancient plants and the world BERTRAM, U. 1989. Untersuchungen an coal balls aus they lived in. Comstock Press, Ithaca, NY. 279 pp. dem Namur A von Ostrau unter spezieller Beriick- ARENS, N. C. 1993. Reconstructing community dy- sichtigung der Gattungen Heterangium, Lyginop- namics in a Pennsylvanian floodplain forest. Ph.D. teris und Microspennopteris. Palaeontographica Thesis, Harvard University, Cambridge, MA. 214B: 125-224. BAKER, R. A. AND W. A. DIMICHELE. 1997. Biomass BERTRAND, P. AND P. CORSIN. 1950. Reconstitution de allocation in Late Pennsylvanian coal-swamp tableaux de vegetation fossile. Annales de Paleon- plants. Palaios 12: 127-132. tologie 61: 126-144. BARTHEL, M. 1962. Epidermisuntersuchungen an ei- BLAKE, B. M., A. T. CROSS, C. E EBLE, W H. GILLES- nigen inkohlten Pteridospermenbliittern des Ober- PIE, AND H. W PFEFFERKORN. 2002. Selected plant karbons und Perms. Geologic, Beiheft 33: 1—140. megafossils from the Carboniferous of the Appa- BARTHEL, M. 1976. Die Rotliegendflora Sachsens. Ab- lachian region, eastern United States: Geographic handlungen Staatliches Museum fiir Geologic und and stratigraphic distribution, p.259—335. In L. V. Mineralogie Dresden 24: 1—190. Hills, C. M. Henderson, and E. W Bamber [eds.]. BARTHEL, M. 2001. Amhardtia scheibei (Gothan) Hau- Carboniferous and Permian of the World. Proceed- bold & Kerp und andere Rotliegend-Peltasperma- ings XIV International Congress of Carboniferous- ceen. Wage und Irrwege ihrer Enforschung. Hal- Permian. Canadian Society of Petroleum Geolo- lesches Jahrbuch fiir Geowissenschaften, B 23: 1—8. gists, Memoir 19. BARTHEL, M. AND R. ROESSLER. 1994. Palaontologische BOERSMA, M. 1972. The heterogeneity of the form ge- Fundschichten im Rotliegend von Manebach (Thiir. nus Mariopteris Zeiller: A comparative morpholog- Wald) mit Calamites gigas (Sphenophyta). Veroff. ical study with special reference to the frond com- Naturhist. Mus. Schleusingen 11: 3—21. position of West-European species. Ph.D. Disser- tation. Drukkerij Elinkwijk, Utrecht, Netherlands. BARTHEL, M. AND R. ROESSLER. 1996. Calamiten im Oberrotliegend des Thiiringer Waldes.—Was ist BROUTIN, J., J. DOUBINGER, J. LANGIAUX, AND D. PRI- "Walchia imbricate"? Veroff. Naturhist. Mus. MLY. 1986. Consequences de la coexistence de flores a caracteres stephaniens et autuniens dan les Schleusingen 9: 69-80. bassins limniques d'Europe occidentale. Memoires BATEMAN, R. M. 1991. Palaeoecology, p. 34-116. In Societe Geologique France 149: 15—25. C. J. Cleal (ed.). Plant Fossils in Geological Inves- BROUTIN, J., J. DOUBINGER, G. FARJANEL, E FREYTET, tigation: the Palaeozoic. Ellis Horwood, Chichester, H. KERP, J. LANGIAUX, L. LEBERTON, S. SEBBAN, UK. AND S., SATTA. 1990. Le renouvellement des flores BATEMAN, R. M. 1992. MOrphometric reconstruction, au passage Carbonifere Permien: approaches stra- paleobiology and phylogeny of Oxroadia gracilis tigraphique, biologique, sedimentologique. Aca- Alvin emend, and O. conferta sp. nov.: Anatomi- deme des Sciences Paris, Comptes Rendues 311: cally preserved rhizomorphic lycopsids from the 1563-1569. Dinantian of Oxroad Bay, SE Scotland. Palaeon- BUISINE, M. 1961. Les Alethopteridees du Nord de la tographica 228: 29-103. France. Etudes Geologiques, Service Geologique BATEMAN, R. M. AND G. W ROTHWELL. 1990. A re- Houilleres du Bassin du Nord et du Pas-de-Calais appraisal of the Dinantian floras at Oxroad Bay, 4: 1-317. East Lothian, Scotland. 1. Floristics and the devel- CECIL, C. B., E T. DULONG, R. R. WEST, R. STAMM, opment of whole-plant concepts. Trans. Roy. Soc. AND B. WARDLAW. 2003. Climate controls on the Edinburgh 81: 127-159. stratigraphy of a Middle Pennsylvanian cyclothem BATEMAN, R. M. AND A. C. SCOTT. 1990. A reappraisal in North America. SEPM Special Publication 77: of the Dinantian floras at Oxroad Bay, East Lothi- 151-180. an, Scotland. 2. Volcanicity, palaeoenvironments CHALOT-PRAT, F AND J. GALTIER. 1989. Decouverte and palaeocology. Trans. Roy. Soc. Edinburgh 81: d'un tronc de gymnosperme dans une coulee du 161-194. complexe volcanique carbonifere du Tazekka (Mar- BATEMAN, R. M. AND W A. DIMICHELE. 1994a. Het- oc oriental) et sa signification paleoecologique. erospory: the most iterative key innovation in the Comptes Rendus de I'Academie des Sciences Paris evolution of plants. Biol. Rev. 69: 345-417. 309: 1735-1741. BATEMAN, R. M. AND W A. DIMICHELE. 1994b. Sal- CLEAL, C. J. AND C. H. SHUTE. 1995. A synopsis of tational evolution of form in vascular plants: a neuropterid foliage from the Carboniferous and neoGoldschmidtian synthesis, p. 61—100. In D. S. Lower Permian of Europe. Bulletin of the Natural Ingram [ed.]. Shape and Form in Plants and Fungi. History Museum, London, Geology Series 51: 1— Symposium Series 16. Linnean Society, London, 52. UK. CLEAL, C. J. AND C. H. SHUTE. 2003. Systematics of BATENBURG, L. H. 1982. "Compression species" and the Late Carboniferous meduUosalean pteridosperm "petrifaction species" of Sphenophyllum com- Laveineopteris and its associated Cyclopteris pared. Rev. Palaeobotany and Palynology 36: 335— leaves. Palaeontology 46: 353—411 359. CLEAL, C. J., R. M. JAMES, AND E. L. ZODROW. 1999. BAXTER, R. W 1949. Some pteridospermous stems and Variation in stomatal density in the Late Carbon- 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 111

iferous gymnosperm frond Neuropteris ovata. Pa- The Nonmarine Permian. New Mexico Museum of laios 14: 180-185. Natural History and Science Bulletin 30: 60—66. CLEAL, C. J., J.-P LAVEINE, AND C. H. SHUTE. 1996. DiMiCHELE, W. A., T L. PHILLIPS, AND G. E. MCBRINN. Further observations on the Upper Carboniferous 1991. Quantitative analysis and paleoecology of the pteridosperm frond Macroneuropteiis macrophylla. Secor coal and roof-shale floras (Middle Pennsyl- Palaeontology 41: 383-386. vanian, Oklahoma). Palaios 6: 390-409. CLEAL, C. J., C. H. SHUTE, AND Z. L. ZODROW. 1990. DiMiCHELE, W. A., W. E. STEIN, AND R. M. BATEMAN. A revised taxonomy for Palaeozoic neuropterid fo- 2001. Ecological sorting during the Paleozoic ra- liage. Taxon 39: 486-492. diation of vascular plant classes, p. 285—235. In W. COTTA, C. B. 1832. Die Dendrolithen in Beziehung auf D. AUmon and D. J. Bottjer [eds.]. Evolutionary ihren innerenBau. 89 pp. Dresden and Leipzig, Paleoecology. Columbia University Press, New Germany. York, NY. CROOKALL, R. 1976. Fossil plants of the Carboniferous DiMiCHELE, W. A., H. KERP, M. KRINGS, AND D. S. rocks of Great Britain (second section). Memoirs CHANEY. 2005b. The Permian peltasperm radiation: of the Geological Survey of Great Britain, Palaeon- evidence from the southwestern United States. In tology 4(Part 7): 841-1004. S. Lucas [ed.]. The Nonmarine Permian. New Mex- DARRAH, W. C. 1969. A critical review of the Upper ico Museum of Natural History and Science Bul- Pennsylvanian floras of eastern United States with letin 30: 67-79. notes on the Maxon Creek Flora of Illinois. Pri- DiMiCHELE, W. A., S. H. MAMAY, D. S. CHANEY, R. vately published by W. C. Darrah. 220 pp. W. HOOK AND W. J. NELSON. 2001. An early Perm- DAVIES, D. 1929. Correlation and palaeontology of the ian flora with late Permian and Mesozoic affinities Coal Measures in East Glamorganshire. Phil. Trans. from north-central Texas. J. of Paleontology 75: Roy. Soc. London 217B: 91-153. 449-460. DEMKO, T. AND R. A. GASTALDO. 1992. Paludal envi- DiMiCHELE, W. A., M. O. RiSCHBIETER, D. L. EGGERT, ronments of the Mary Lee coal zone, Pottsville For- AND R. A. GASTALDO. 1984. Stem and leaf cuticle mation, Alabama: stacked clastic swamps and peat of Karinopteiis: source of the cuticles from the In- mires. Int. J. Coal Geol. 20: 23-47. diana "paper" coal. Am. J. Bot. 71: 626—637. TABOR, D. S. CHANEY, AND DENNIS, R. L. AND D. A. EGGERT. 1978. Parasporo- DiMiCHELE, W. A., N. J. theca, gen. nov., and its bearing on the interpreta- W. J. NELSON. 2006. From wetlands to wetspots: tion of the morphology of permineralized medul- the fate and significance of Carboniferous elements losan pollen organs. Bot. Gaz. 139: 117—139. in Early Permian coastal plain floras of North-Cen- tral Texas. In S. Greb and W. A. DiMichele [eds.]. DiMiCHELE, W. A. AND R. B. ARONSON. 1992. The Wetlands Through Time. Geological Society of Pennsylvanian-Permian vegetational transition: a America Special Publication. terrestrial analogue to the onshore-offshore hypoth- DiMiTROVA T KH., C. J. CLEAL, AND B. A. THOMAS. In esis. Evolution 46: 807-824. press. Palynology of late Westphalian-early Ste- DiMiCHELE, W. A. AND R. M. BATEMAN. 1996. Plant phanian coal-bearing deposits in the eastern South paleoecology and evolutionary inference: two ex- Wales Coalfield. Geological Magazine. amples from the Paleozoic. Rev. Paleobot. Palynol. DOYLE, J. A. 1996. Seed plant phylogeny and the re- 90: 223-247. lationships of Gnetales. Int. J. Plant Sci. 157(sup- DiMiCHELE, W. A. AND W. J. NELSON. 1989. Small- plement): S3-S39. scale spatial heterogeneity in Pennsylvanian-age DRAGERT, K. 1964. Pflanzensoziologische Untersu- vegetation from the roof-shale of the Springfield chungen in den Mittleren Essener Schichten des Coal. Palaios 4: 276-280. nordlichen Ruhrgebietes. Forschungsberichte Land DiMiCHELE, W. A. AND T L. PHILLIPS. 1988. Paleo- Nordrhein-Westfalen 1363: 1-111. ecology of the Middle Pennsylvanian-age Herrin DUFEK, D. AND B. M. STIDD. 1981. The vascular sys- coal swamp near a contemporaneous river system, tem of Dolerotheca and its phylogenetic signifi- the Walshville Paleochannel. Rev. Palaeobot. Pa- cance. Am. J Bot. 68: 897-907. lynol. 56: 151-176. DUNN, M. T 2004. The Fayetteville flora I: Upper Mis- DiMiCHELE, W. A. AND T. L. PHILLIPS. 1996a. Clades, sissippian (middle Chesterian/lower Namurian A) ecological amplitudes, and ecomorphs: phylogenet- plant assemblages of the permineralized and com- ic effects and the persistence of primitive plant pression remains from Arkansas, USA. Rev. Pa- communities in the Pennsylvanian-age tropics. Pa- laeobot. Palynol. 132: 79-102. laeogeography, Palaeoclimatology, Palaeoecology DUNN, M. T, G. MAPES, AND G. W. ROTHWELL. 2002a. 127: 83-106. On Paleozoic plants from marine strata: Hexaloba DiMiCHELE, W. A. AND T. L. PHILLIPS. 1996b. Climate finisensia new genus and species, a trigonocarpa- change, plant extinctions, and vegetational recov- lean ovule from the Virgilian (Upper Pennsylva- ery during the Middle-Late Pennsylvanian transi- nian: Gzhelian) Finis Shale of Texas. J. Paleontol- tion: the case of tropical peat-forming environ- ogy 76: 173-180. ments in North America, p. 201—221. /« M. L. Hart DUNN, M. T, G. W. ROTHWELL, AND G. MAPES. 2002b. [ed.], Biotic Recovery from Mass Extinctions. Geo- Additional observations on Rhynchosperma quinnii logical Society Special Publication 102. London, (MeduUosaceae): a permineralized ovule from the UK. Chesterian (Upper Mississippian) Fayetteville For- DiMiCHELE, W. A., N. J. TABOR, AND D. S. CHANEY. mation or Arkansas. Am. J. Bot. 89: 1799-1808. 2005b. Outcrop-scale environmental heterogeneity DUNN, M. T, G. W. ROTHWELL, AND G. MAPES. 2003a. and vegetational complexity in the Permo-Carbon- On Paleozoic plants from marine strata: Tiivena ar- iferous Markley Formation of North Central Texas. kansana (Lyginopteridaceae) gen. et sp. nov., a ly- 112 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

ginopterid plant from the Fayetteville Formation sperms. Courier Forschungsinstitut Senckenberg (middle ChesterianAJpper Mississippian) of Arkan- 147: 119-125. sas, USA. Am. J. Bot. 90: 1239-1252. GALTIER, J. AND O. BETHOUX, O. 2002. Morphology DUNN, M. T, G. W. ROTHWELL, AND G. MAPES. In and growth habit of Dicksonites pluckenetii from press. The Fayetteville flora of Arkansas (USA): a the Upper Carboniferous of Graissessac (France). snapshot of terrestrial vegetation patterns within a Geobios 35: 525—535. clastic swamp at Late Mississippian time. In S. GALTIER, J. AND J. BROUTIN. 1995. La flore de la Zone Greb and W. A. DiMichele [eds.]. Wetlands de Transition de I'Autunien superieur de Lodeve. Through Time. Geological Society of America Spe- Sciences Geologiques Bulletin, Strasbourg 48: 83— cial Paper. 93. DUNN, M. T, M. KRINGS, G. MAPES, G. W. ROTHWELL, GALTIER, J., R. E. BROWN, A. C. SCOTT, G. M. REX, R. H. MAPES, AND SUN KEQIN. 2003b. Medullosa AND N. P. ROWE. 1993. A Late Dinantian flora from steinii sp. nov., a seed fern vine from the Upper Weaklaw, East Lothian, Scotland. Special Papers in Mississippian. Rev. Palaeobot. Palynol. 124: 307— Palaeontology 49: 57-74. 324. GALTIER, J., B. MEYER-BERTHAUD, AND C. B. BECK. EBLE, C. F AND S. F GREB. 2004. Palynostratigraphy 1993. Large Calamopitys stems from the Tourna- and paleoecology of a vertebrate locality at the sian of France. Palaeontographica B 230: 59—79. Mississippian-Pennsylvanian boundary in western GALTIER, J., B. MEYER-BERTHAUD, AND N. P. ROWE. Kentucky. Geological Society of America Annual 1988. Tournaisian plants from the "Lydiennes" Meeting (Denver), Abstracts with Programs 36(5): Formation of the Montagne Noire (France). Courier 84. Forschungsinstitut Senckenberg 100: 109—117. ERWIN, D. M., H. W. PFEFFERKORN, AND V. ALLEMAN. GALTIER, J. AND A. C. SCOTT. 1990. On Eristophyton 1994. Early seed plants in the Southern Hemi- and other gymnosperms from the Lower Carbon- sphere: I. Associated ovulate and microsporangiate iferous of Castleton Bay, East Lothian, Scotland. organs from the Carboniferous of Peru. Rev. Paleo- Geobios 23: 5-19. bot. Palynol. 80: 19-38. GALTIER, J. AND A. C. SCOTT. 1994. Arborescent gym- EVANS, B., C. CLEAL, B. THOMAS, AND W. P. A. WIM- nosperms from the Visean of East Kirkton, West BLEDON. 2003. The Westphalian succession of the Lothian, Scotland. Trans. Roy. Soc. Edinburgh, Glyn-neath area. South Wales: an internationally Earth Sciences 84: 261—266. important geological resource. Proc. Geolog. As- GASTALDO, R. A. 1985. Upper Carboniferous paleo- soc. 114: 291-305. ecological reconstructions: observations and recon- FAIRON-DEMARET, M. AND S. E. SCHECKLER. 1987. siderations, p. 281—296. In J. L. Escobeda, L. F Typification and redescription of Moresnetia zales- Granados, B. Melendez, R. Pignatelli, R. Ray and skyi Stockmans, 1948, an early seed plant from the R. H. Wagner [eds.], Dixieme Congres Internation- Upper Famennian of Belgium. Bull. Inst. Roy. Sci. al de Stratigraphie et de Geologic du Carbonifere, Nat. Belg., Sci. de la Terre 57: 183-199. Madrid (1983). FALCON-LANG, H. 2003. Response of Late Carbonif- GASTALDO, R. A. 1986. Implications on the paleoecol- erous tropical vegetation to transgressive-regressive ogy of autochthonous lycopods in clastic sedimen- rhythms at Joggins, Nova Scotia. J. Geol. Soc. Lon- tary environments of the Early Pennsylvanian of don 160: 643-648. Alabama. Palaeogeography, Palaeoclimatology, Pa- FALCON-LANG, H. 2004. Pennsylvanian tropical rain laeoecology 53: 191—212. forests responded to glacial-interglacial rhythms. GASTALDO, R. A. 1987. Confirmation of Carboniferous Geology 32: 698-692. clastic swamp communities. Nature 326: 869—871. FALCON-LANG, H. In press. Vegetation of Early Penn- GASTALDO, R. A. 1988. The frond architecture of sylvanian alluvial fan and piedmont environments Sphenopteris pottsvillea (White) Gastaldo and in southern New Brunswick, Canada. Palaeogeog- Boersma. J. Paleontology 62: 982-991. raphy, Palaeoclimatology, Palaeoecology. GASTALDO, R. A. AND M. A. BOERSMA. 1983a. A re- FAVRE-DUCHARTRE, M. 1956. Contribution a I'etude de investigation of Early Pennsylvanian species of la reproduction chez le Ginkgo biloba. Rev. Cytol. Mariopteris from the Appalachian Region. I. Kar- Biol. Veg. 17: 1-218. inopteris, Mariopteris and the 'Pottsvillea Com- FRANKS, J. W 1963. On some alethopterid pinnules plex.' Rev. Palaeobot. Palynol. 38: 185-226. from the Permian of Autun. Ann. Bot., n.s. 27: GASTALDO, R. A. AND M. A. BOERSMA. 1983b. A re- 447-452. investigation of Early Pennsylvanian species of GALTIER, J. 1974. Sur I'organisation de la fronde des Mariopteris from the Appalachian Region. II. Eus- Calamopitys. Pteridospermales probables du Car- phenopteris and Sphenopteris. Rev. Palaeobot. Pa- bonifere inferieur. Comptes Rendus de I'Academie lynol. 38: 227-247. des Sciences 279: 975-978. GASTALDO, R. A., W A. DIMICHELE, AND H. W PFEF- GALTIER, J. 1986. Morphologic vegetative des progym- FERKORN. 1996. Out of the icehouse into the green- nosperms et Pteridospermales dy Paleozoique. Na- house: a late Paleozoic analogue for modern global turalia monspeliensia—CoUoque international sur vegetational change. GSA Today 6: 1—7. I'Arbre. p. 495-505. GASTALDO, R. A. AND L. C. MATTEN. 1978. Trigono- GALTIER, J. 1988. Morphology and phylogeny of early carpus leanus, a new species from the Middle pteridosperms, p. 135—176. In C. B. Beck [ed.]. Or- Pennsylvanian of southern Illinois. Am. J. Bot. 65: igin and Evolution of Gymnosperms. Columbia 882-890. University Press, New York, NY. GASTALDO, R. A., I. STEPANOVIC-WALLS, W N. WARE, GALTIER, J. 1992. On the earliest arborescent gymno- AND S. F. GREB. 2004. Community heterogeneity of 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 113

Early Pennsylvanian peat mires. Geology 32: 693— Population Biology 32. Princeton University Press, 696. Princeton, NJ. GERRIENNE, P., M. FAIRON-DEMARET, AND J. GALTIER. lANNUZZi, R. AND H. W PFEFFERKORN. 2002. A pre- 1999. A Namurian A (Silesian) permineralized flo- glacial, warm-temperate floral belt in Gondwana ra from the Carriere du Lion at Engihoul (Bel- (Late Visean, Early Carboniferous). Palaios 17: gium). Rev. Palaeobot. Palynol. 107: 1—15. 571-590. GERRIENNE, P., B. MEYER-BERTHAUD, M. FAIRON-DE- JENNINGS, J. R. 1976. The morphology and relation- MARET, M. STREEL AND P. STEEMANS. 2004. Run- ships of Rhodea, Telangium, Telangiopsis, and caria, a Middle Devonian seed plant precursor. Sci- Heterangium. Am. J. Bot. 63: 1119-1133. ence 306: 856-858. KERP, J. H. F 1982. Aspects of Permian palaeobotany GILINSKY, N. L. AND R. K. BAMBACH. 1987. Asym- and palynology. XI. On the presence of the ovuli- metrical patterns of origination and extinction in ferous organ Autunia niilleiyensis (Renault) Kras- higher taxa. Paleobiology 13: 427—445. ser (Peltaspermaceae) in the Lower Permian of the GiLLESPiE, W. H. AND H. W. PFEFFERKORN. 1979. Dis- Nahe area and its relationship to Callipteris con- tribution of commonly occurring plant megafossils ferta (Stemberg) Brongniart. Acta Bot. Neerl. 31: in the proposed Pennsylvanian System Stratotype, 417-427. p.87—96. In K. J. Englund et al. [eds.]. Proposed KERP, J. H. F 1988. Aspects of Permian palaeobotany Pennsylvanian System Stratotype Virginia and and palynology. X. The West- and Central Euro- West Virginia. 9* International Congress of Car- pean species of the genus Autunia Krasser emend. boniferous Stratigraphy and Geology, American (Peltaspermaceae) and the form-genus Rhachiphyl- Geological Institute, Selected Guidebook Series 1. luni Kerp (callipterid foliage). Rev. Palaeobot. Pa- GiLLESPiE, W. H., G. W. ROTHWELL, AND S. E. SCHECK- lynol. 54: 249-360. LER. 1981. The earliest seeds. Nature 293: 462- KERP, H. 1996. Post-Variscan late Palaeozoic Northern 464. Hemisphere gymnosperms: the onset to the Meso- GORDON, W. T. 1912. On Rhetinangium arberi, a new zoic. Rev. Palaeobot. Palynol. 90: 263-285. genus of Cycadofilices from the Calciferous Sand- KERP, H. 2000. The modernization of landscapes dur- stone Series. Trans. Roy. Soc. Edinburgh 48: 813— ing the late Paleozoic—Early Mesozoic. In R. A. 825. Gastaldo and W A. DiMichele [eds.], Phanerozoic GRAND'EURY, F C. 1905. Sur les graines trouvees at- Terrestrial Ecosystems. Paleontological Society tachees au Pecopteris Pluckeneti Schlot. Compte Special Papers. 6: 79—113. Rendus de I'Academie des Sciences de Paris 141: KERP, H. AND M. BARTHEL. 1993. Problems of cutic- 920-923. ular analysis of pteridosperms. Rev. Palaeobot. Pa- HALLE, T G. 1927. Palaeozoic plants from central lynol. 78: 1-18. Shansi. Paleontologia Sinica, Series A, 11, Fascicle KERP, J. H. F AND J. FIGHTER. 1985. Die Makrofloren 1, 316 pp. des saarpfalzischen Rotliegenden (Ober-Karbon- HAMER, J. J. AND G. W. ROTHWELL. 1988. The vege- Unter-Perm; SW Deutschland). Mainzer Geowis- tative structure of Medullosa endocentiica (Pteri- senschaftliche Mitteilungen 14: 159-286. dospermopsida). Can. J. Bot 66: 375—387. KERP, J. H. F AND H. HAUBOLD. 1988. Aspects of HARTUNG, W. 1938. Flora und Altersstellung des Kar- Permian palaeobotany and palynology. VIII. On the bons von Hainischen-Ebersdorf und Bomia bei reclassification of the West- and Central European Chemnitz (erzgeb. Becken). Abhundlungen des species of the form-genus Callipteris Brongniart Siichsischen Geologischen Landesamtes 18: 1—140. 1849. Rev. Palaeobot. Palynol. 54: 135-150. HiCKEY, L. J., AND J. A. DOYLE. 1977. Early Creta- KERP, H. AND M. KRINGS. 1998. Climbing and scram- ceous fossil evidence for angiosperm evolution. bling growth habits: common life strategies among Bot. Rev. 43: 1-104. Late Carboniferous seed ferns. Compte Rendu, HILTON, J. M. AND R. M. BATEMAN. 2006. Pterido- Academic des Sciences, Paris, Earth and Planetary sperms are the backbone of seed-plant phylogeny. Sciences 326: 583-588. Journal of the Torrey Botanical Society 133: 119— KERP, H. AND M. KRINGS. 2003. Kutikularanalytische 168. Befunde zur Abrenzung der Odontopteris minor f. HILTON, J. M. AND D. EDWARDS. 1996. A new Late zeilleri Potonie von Odontopteris reichiana Gut- Devonian acupulate preovule from Taff George, bier Courier Forschungsinstitut Senckenberg 241: South Wales. Rev. Palaeobot. Palynol. 93: 235- 31-42. 252. KERP, H., J. BROUTIN, S. LAUSBERG, AND H. AASSOUMI. HosKiNS, J. H. AND A. T CROSS. 1946. Studies in the 2001. Discovery of Latest Carboniferous-Early Trigonocarpales. Part II. Taxonomic problems and Permian radially symmetrical peltaspermaceous a revision of the genus Pachytesta. Am. Midi. Nat. megasporophylls from Europe and North Africa. 36: 331-361. Comptes Rendus de I'Academie des Sciences de HOTTON, C. L. AND W E. STEIN. 1994. An ontogenetic Paris, Serie Ila 322: 513-519. model for the Mississippian seed plant family Cal- KNOLL, A. H., K. J. NIKLAS, P. G. GENSEL, AND B. H. amopityaceae. Int. J. Plant Sci. 155: 119-142. TiFFNEY. 1984. Character diversification and pat- HoTTON, N., E. C. OLSON, AND R. BEERBOWER. 1997. terns of evolution in early vascular plants. Paleo- Amniote origins and the discovery of herbivory, biology 10: 34-47. pp. 207-264. In S. S. Sumida and K. L. M. Martin KRINGS, M. AND H. KERP. 1997. Cuticles of Lescurop- [eds.], Amniote Origins. Academic Press, San Di- teris genuina from the Stephanian (Upper Carbon- ego, CA. iferous) of Central France: evidence for a climbing HUBBELL, S. P 2001. The Unified Neutral Theory of growth habit. Bot. J. Linnean Soc. 123: 73-89. Biodiversity and Biogeography. Monographs in KRINGS, M. AND H. KERP. 1998. Epidermal anatomy of 114 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

Bai-thelopteris germaiii from the Upper Carbonif- from architecture of the Carboniferous pterido- erous and Lower Permian of France and Germany. sperm Laveineopteris rarinervis, with the support Am. J. Bot. 85: 553-562. of some recently discovered specimens. Revue de KRINGS, M. AND H. KERP. 1999. Morphology, growth Paleobiologie, Geneve. 17: 381—443. habit, and ecology of Blanzyopteris praedentata LAVEINE, J. P., Y. LEMOIGNE, AND ZHANG SHANZHEN. (Gothan) nov. comb., a climbing neuropterid seed 1993. General characteristics and paleobiogeogra- fern from the Stephanian of Central France. Int. J. phy of the Parispermaceae (genera Paripteris Goth- Plant Sci. 160: 603-619. an and Linopteris Presl), pteridosperms from the KRINGS, M. AND H. KERP. 2000. A contribution to the Carboniferous. Palaeontographica B 230: 81—139. knowledge of the pteridosperm genera Pseudoma- LAVEINE, J.-P, ZHANG SHANZHEN, AND Y LEMOIGNE. liopteris Danze-Corsin nov. emend. And Helen- 1997. L'importance de I'approche qualitative en pa- opteris nov. gen. Rev. Palaeobot. Palynol. Ill: leontologie: I'exemple des "Neuropteridees" (Pter- 145-196. idospermees du Carbonifere). Quelques implica- KRINGS, M., T. N. TAYLOR AND D. W. KELLOGG. 2002. tions. Geobios 20: 315-329. Touch-sensitive trichomes: a mode of defence LEPAGE, B. A. AND H. W. PFEFFERKORN. 2000. Did against herbivorous arthropods in the Carbonifer- ground cover change over geologic time?, p. 171— ous. Evol. Ecol. Res. 4: 779-786. 182. In R. A. Gastaldo and W. A. DiMichele [eds.], KRINGS, M., D. W. KELLOGG, H. KERP, AND T N. TAY- Phanerozoic Terrestrial Ecosystems. Paleontologi- LOR. 2003a. Trichomes of the seed fern Blanzyop- cal Society Special Papers 6. teris praedentata: implications for plant-insect in- LONG, A. G. 1960a. Stamnostoma huttonense gen. et teractions in the Late Carboniferous. Bot. J. Lin- sp. nov.—A pteridosperm seed and cupule from the nean Soc. 141: 133-149. Calciferous Sandstone Series of Berwickshire. KRINGS, M., H. KERP, E. L. TAYLOR, AND T N. TAYLOR. Trans. Roy. Soc. Edinburgh 64: 201—215. 2001. Reconstruction of Pseudomariopteiis bus- LONG, A. G. 1960b. On the structure of Samaropsis quetii, a vine-like Late Carboniferous-Early Perm- scotica Calder (emended) and Eurystoma angulare ian pteridosperm. Am. J. Bot 88: 161—116. gen. et sp. nov., petrified seeds from the Calciferous KRINGS, M., H. KERP, T N. TAYLOR, AND E. L. TAYLOR. Sandstone Series of Berwickshire. Trans. Roy. Soc. 2003b. How Paleozoic vines and lianas got off the Edinburgh 64: 261-280. ground: on scrambling and climbing Carbonifer- LONG, A. G. 1961a. On the structure of Deltasperma ous-Early Permian pteridosperms. Bot. Rev. 69: fouldenense gen. et sp. nov., and Camptosperma 204-244. berniciense gen. et sp. nov., petrified seeds from KRINGS, M., S. D. KLAVINS, W. A. DIMICHELE, H. the Calciferous Sandstone Series of Berwickshire. KERP, AND T N. TAYLOR. 2005. Epidermal anatomy Trans. Roy. Soc. Edinburgh 64: 281-295. of Glenopteris splendens nov. emend., an enigmatic LONG, A. G. 1961b. Some pteridosperm seeds from the seed plant from the Lower Permian of Kansas Calciferous Sandstone Series of Berwickshire. (U.S.A.). Rev. Palaeobot. Palynol. 136: 159-180. Trans. Roy. Soc. Edinburgh 64: 401-419. KRINGS, M., T N. TAYLOR, E. L. TAYLOR, B. J. LONG, A. G. 1961c. Tristichia ovensi gen. et sp. nov., AxsMiTH. AND H. KERP. 2001. Cuticles of Maiiop- teris occidentalis White nov. emend, from the Mid- a protostelic Lower Carboniferous pteridosperm dle Pennsylvanian of Oklahoma (USA), and a new from Berwickshire and East Lothian, with an ac- type of climber hook for the mariopterid pterido- count of some associated seeds and cupules. Trans. sperms. Rev. Palaeobot. Palynol. 114: 209—222. Roy. Soc. Edinburgh 64: 477-489. LABANDEIRA, C. C. 2000. The paleobiology of polli- LONG, A. G. 1964. Some specimens of Stenomyelon nation and its precursors, p. 233—269. In R. A. Gas- and Kalymma from the Calciferous Sandstone Se- taldo and W. A. DiMichele [eds.]. The Origin and ries of Berwickshire. Trans. Roy. Soc. Edinburgh Development of Terrestrial Ecosystems. Paleonto- 65: 435-447. logical Society Short Course 6. LONG, A. G. 1965. On the cupule structure of Eurys- LAMBOY, W. AND A. LESNIKOWSKA. 1988. Some statis- toma angulare. Trans. Roy. Soc. Edinburgh 66: tical methods useful in the analysis of plant paleo- 111-128. ecological data. Palaios 3: 86—94. LONG, A. G. 1969. Eurystoma trigona sp. nov., a pte- LANGIAUX, J. 1986. Decouverte d'un specimen excep- ridosperm ovule borne on a frond of Alcicomop- tionnel de fructification femelle de Dicksonites teris Kidston. Trans. Roy. Soc. Edinburgh 68: 173— sterzeli Zeiller 1888 dans le Stephanien du Massif 182. Central fran9ais. Comptes Rendus de I'Academie LONG, A. G. 1975. Further observations on some Low- des Sciences Paris 302, Serie II 11: 755—760. er Carboniferous seeds and cupules. Trans. Roy. LAVEINE, J.-P. 1986. The size of the frond in the genus Soc. Edinburgh 69: 267-293. Alethopteris Sternberg (Pteridospermopsida, Car- LONG, A. G. 1976. Calathopteris heterophylla gen. et boniferous). Geobios 19: 49—56. sp. nov., a Lower Carboniferous pteridosperm bear- LAVEINE, J.-P. 1997. Synthetic analysis of the Neurop- ing two kinds of petioles. Trans. Roy. Soc. Edin- terids. Their interest for the decipherment of Car- burgh 69: 327-336. boniferous palaeogeography. Rev. Palaeobot. Pa- LONG, A. G. 1977a. Some Lower Carboniferous pte- lynol. 95: 155-189. ridosperm cupules bearing ovules and microsporan- LAVEINE, J.-R, R. COQUEL, AND S. LOBOZIAK. 1977. gia. Trans. Roy. Soc. Edinburgh 70: 1—11. Phylogenie generale des Callipteridiacees (Pteri- LONG, A. G. 1977b. Lower Carboniferous pterido- dospermopsida). Geobios 10: 757—847. sperm cupules and the origin of angiosperms. LAVEINE, J.-P, L., AND H. DUQUESNE. 1998. Morpho- Trans. Roy. Soc. Edinburgh 70: 13—35. logical analysis of the "bifurcate semipinnate" LONG, A. G. 1977c. Observations on Carboniferous 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 115

seeds of Mitrospermum^ Conostoma and Lagenos- NiKLAS, K. J. 1981b. Air flow patterns around some tonia. Trans. Roy. Soc. Edinburgh 70: 37—61. early seed plant ovules and cupules: implications LONG, A. G. 1979a. Observations on the Lower Car- concerning efficiency in wind pollination. Am. J. boniferous genus Pitus Witham. Trans. Roy. Soc. Bot 68: 635-650. Edinburgh 70: 111-127. NIXON, K. C, W. L. CREPET, D. W. STEVENSON, AND LONG, A. G. 1979b. The resemblance between the E. M. FRIIS. 1994. A reevalution of seed plant phy- Lower Carboniferous cupules of Hydraspemia cf. logeny. Ann. Missouri Bot. Garden 81: 484-533. tenuis Long and Sphenopteiis bifida Lindley and OESTRY-STIDD, L. L. 1979. Anatomically preserved Hutton. Trans. Roy. Soc. Edinburgh 70: 129-137. Neuropteris rarineivis from American coal balls. J. LONG, A. G. 1987. Observations on Eristophyton Za- Paleontology 53: 37-43. lessky, Lyginorachis waltoni Calder, and Cladoxy- OLIVER, F W. AND D. H. SCOTT. 1904. On the structure lon edromense sp. nov. from the Lower Carbonif- of the Palaeozoic seed Lagenostoma lomaxi, with erous Cementstone Group of Scotland. Trans. Roy. a statement of the evidence upon which it is re- Soc. Edinburgh 78: 73-84. ferred to Lyginodendron. Phil. Trans. Royal Soc. LYONS, P. C. AND W. C. DARRAH. 1989. Earliest coni- London, Series B 231: 193-247. fers in North America: Upland and/or paleoclimatic PATTEISKY, K. 1957. Die phylogenetische Entwicklung indicators. Palaios 4: 480-486. der Arten von Lyginopteris und ihre Bedeutung fiir MACK, G. H., M. LEEDER, M. PEREZ-ARLUCEA, AND B. die Stratigraphie. Mitteilungen der westfiilischen D. J. BAILEY. 2002. Early Permian silt-bed fluvial Berggewerkschaftskasse, Bochum 12: 59—83. sedimentation in the Orogrande basin of the An- PEPPERS, R. A. 1996. Palynological correlation of ma- cestral Rocky Mountains, New Mexico, USA. Sed- jor Pennsylvanian (Middle and Upper Carbonifer- imentary Geology 160: 159—178. ous) chronostratigraphic boundaries in the Illinois MACK, G. H., M. LEEDER, M. PEREZ-ARLUCEA, AND B. and other coal basins. Geol. Soc. Amer Mem. 188. D. J. BAILEY. 2003. Sedimentology, paleontology, Boulder, CO. 111 p. and sequence stratigraphy of Early Permian estua- PFEFFERKORN, H. W. AND W. H. GILLESPIE. 1982. Plant rine deposits, south-central New Mexico, USA. Pa- megafossils near the Mississippian-Pennsylvanian laios 18: 403-420. boundary in the Pennsylvanian System Stratotype, MACK, G. H., D. R. COLE, T H. GIORDANO, W. C. West VirginiaA'irginia, p. 128-133. In W. H. C. SCHAAL, AND J. H. BARCELOS. 1991. Paleoclimatic Ramsbottom, W. B. Saunders, and B. Owens [eds.], controls on stable oxygen and carbon isotopes in Biostratigraphic Data for a Mid-Carboniferous caliche of the Abo Formation (Permian), south-cen- Boundary. Palaeobotanical Contributions of the tral New Mexico, U.S.A. Journal of Sedimentary Subcommission on Carboniferous Stratigraphy, Petrology 61: 458-472. Leeds, UK. MAMAY, S. H. 1992. Sphenopteridium and Telangiop- PFEFFERKORN, H. W. AND M. THOMSON. 1982. Changes sis in a Diplopteiidium-\i]Paleobotany of permineralized peat (coal balls) cal ovuliferous discs from the Kungurian (upper- from the Herrin Coal Member of the Illinois Basin, most Lower Permian) of the Fore-Urals (Russia). In P. H. Given and A. D. Cohen [eds.]. Interdisci- Rev. Palaeobot. Palynol. 91: 35-62. plinary Studies of Peat and Coal. Geological So- NELSON, W. J., R. W. HOOK, AND N. J. TABOR. 2001. ciety of America Microform Publication 7: 18—49. Clear Fork Group (Leonardian, Lower Permian) of PHILLIPS, T L. AND R. A. PEPPERS. 1984. Changing North-Central Texas. Oklahoma Geological Survey patterns of Pennsylvanian coal-swamp vegetation Circular 104: 167-169. and implications of climatic control on coal occur- NiKLAS, K. J. 1981a. Simulated wind pollination and rence. Int. J. Coal Geology 3: 205—255. airflow around ovules of some early seed plants. PHILLIPS, T L., R. A. PEPPERS, M. J. AVCIN, AND P. E Science 211: 275-277. LAUGHNAN. 1974. Fossil plants and coal: patterns 116 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

of change in Pennsylvanian coal swamps of the Il- ROTHWELL, G. W. AND D. A. EGGERT. 1986. A mono- linois Basin. Science 184: 1367—1369. graph of Dolerotheca Halle, and related complex PHILLIPS, T. L., R. A. PEPPERS, AND W. A. DIMICHELE. permineralised meduUosan pollen organs. Trans. 1985. Stratigraphic and interregional changes in Roy. Soc. Edinburgh: Earth Sciences 77: 47—79. Pennsylvanian-age coal-swamp vegetation: envi- ROTHWELL, G. W AND D. M. ERWIN. 1987. Origin of ronmental inferences. Int. J. Coal Geology 5: 43— seed plants: an aneurophyte/seed-fern link elabo- 109. rated. Am. J. Bot. 74: 970-973. PiGG, K. B., R. A. STOCKEY, AND T N. TAYLOR. 1986. ROTHWELL, G. W AND A. C. ScoTT. 1992. Stamnos- Studies of Paleozoic seed ferns: Additional studies toma oliveri, a gymnosperm with systems of ovu- of Microspermopteris aphyllum Baxter. Bot. Gaz. late cupules from the Lower Carboniferous (Din- 147: 126-136 antian) floras at Oxroad Bay, East Lothian, Scot- PiGG, K. B., T. N. TAYLOR, AND R. A. STOCKEY. 1987. land. Rev. Palaeobot. Palynol. 72: 273-284. Paleozoic seed ferns: Heterangium kentuckyensis ROTHWELL, G. W AND R. SERBET. 1992. Pollination sp. nov., from the upper Carboniferous of North biology of Elkinsia polymorpha, implications for America. Am. J. Bot 74: 1184-1204. the origin of gymnosperms. Courier Forschungsin- POTONIE, H. 1893. Die Flora des Rothliegenden von stitut Senckenberg 147: 225-231. Thuringen. Abhandlungen der Koniglishen Preus- ROTHWELL, G. W AND R. SERBET. 1994. Lignophyte sischen Geologischen Landesanstaltn, neue Folge phylogeny and the evolution of the spermato- 9: 1-298. phytes: a numerical cladistic analysis. Syst. Bot. PRINZING, A., W. DURKA, S. KLOTZ, AND R. BRANDL. 19: 443-482. 2001. The niche of higher plants: evidence for phy- ROTHWELL, G. W AND S. E. ScHECKLER. 1988. Biology logenetic conservatism. Proc. Roy. Soc. London B of ancestral gymnosperms, p. 85—134. In C. B. 268: 2383-89. Beck [ed.]. Origin and Evolution of Gymnosperms. PRYOR, J. S. AND R. A. GASTALDO. 2000. Paleoecolog- Columbia University Press, New York, NY. ical analysis of two Early Pennsylvanian mineral- ROTHWELL, G. W AND T N. TAYLOR. 1972. Carbonif- substrate wetlands. Palaios 15: 3—13. erous pteridosperm studies: morphology and anat- PURKYNOVA, E. 1970. Die Untemamurflora des Beck- omy of Schopfiastrum decussatum. Can. J. Bot 50: ens von Homi Slessko (CSSR). Paliiontologie. Ab- 2649-2658. handlungen B., Palaobot. 3: 129-269. ROTHWELL, G. W AND K. L. WHITESIDE. 1974. Rooting RAYMOND, A. 1985. Floral diversity, phytogeography, structures of the Carboniferous meduUosan pteri- and climatic amelioration during the early Carbon- dosperms. Can. J. Bot. 52: 97-102. iferous (Dinantian). Paleobiology 11: 293—309. RowE, N. P, T SPECK, AND J. GALTIER. 1993. Biome- READ, C. B. AND S. H. MAMAY. 1964. Upper Paleozoic chanical analysis of a Palaeozoic gymnosperm flora zones and flora provinces of the United States. stem. Proc. Roy. Soc. London 252: 19—28. U.S. Geological Survey Professional Paper 454-K: ROYER, D. L., L. J. HiCKEY, AND S. L. WiNG. 2003. 1-35. Ecological conservatism in the "living fossil" REIHMAN, M. A. AND J. T SCHABILION. 1976. Cuticles Ginkgo. Paleobiology 29: 84-104. of two species of Alethopteris. Am. J. Bot. 63: SCHABILION, J. T AND M. A. REIHMAN. 1985. Petrified 1039-1046. Neuropteris scheuchzeri pinnules from the Middle REIHMAN, M. A. AND J. T. SCHABILION. 1978. Petrified Pennsylvanian of Iowa: a paleoecological interpre- neuropterid foliage from a Middle Pennsylvanian tation. Compte Rendu, 9th International Carbonif- coal baU. Am. J. Bot. 65: 834-844. erous Congress 5: 3—12. REMY, W. 1975. The floral changes at the Carbonifer- ScHECKLER, S. E. 1986a. Floras of the Devonian-Mis- ous-Permian boundary in Europe and North Amer- sissippian transition, p. 81—96. In R. A. Gastaldo ica, pp. 305—344. In J. A. Barlow [ed.]. Proceed- [ed.]. Land Plants: Notes for a Short Course. Uni- ings of the 1st. I. C. White Memorial Symposium versity of Tennessee, Department of Geological "The age of the Dunkard", West Virginia Geologic Sciences, Studies in Geology 15. and Economic Survey, Morgantown, WV SCHECKLER, S. E. 1986b. Geology, floristics and paleo- REMY, W., R. REMY, G. A. LEISMAN, AND H. HAAS. ecology of Late Devonian coal swamps from Ap- 1980. Der Nachweis von Callipteris flabellifera plachian Laurentia (U.S.A.). Annales de la Societe (Weiss 1897) Zeiller 1898 in Kansas, U.S.A. Ar- geologique de Belgique 109: 209—222. gumenta Palaeobotanica 6: 1—36. SCHOPF, J. M. 1939. MeduUosa distelica, a new species RENAULT, B. 1896. Etude des gites mineraux de la of the anglica group of Medullosa. Am. J. Bot. 26: France: Bassin houiller et permien d'Autun et 196-207. d'Epinac. IV—Flore fossile, Texte. Paris: Impri- SCHULTKA, S. 1995. Die Flora des Namur B in der merie Nationale. 578 pp. ehemalingen Ziegeleitongrube VoBacker bei Fron- ROTHWELL, G. W 1972. Evidence of pollen tubes in denberg: Eine Ubersicht. Geologic und Palaonto- Paleozoic pteridosperms. Science 175: 772—774. logie in Westfalen 35: 3—97. RoTHWELL, G. W 1981. The Callistophytales (Pteri- SCHWENDEMANN, A. B., G., WANG J. M. OSBORN, AND dospermopsida): Reproductively sophisticated Pa- S. L. THATCHER. 2005. Aerodynamics of saccate leozoic gymnosperms. Rev. Palaeobot. Palynol. 32: pollen and its implications for wind pollination. 103-121. Abstracts, Botanical Society of America, Annual RoTHWELL, G. W 1986. Classifying the earliest gym- Meeting, p. 86. nosperms, p. 137—161. In R. A. Spicer and B. A. SCOTT A. C. 1977. A review of the ecology of Car- Thomas [eds.]. Systematic and taxonomic ap- boniferous plant assemblages with new data from proaches in palaeobotany. Systematics Association Strathclyde. Palaeontology 20: 447-73. Special Volume 31. SCOTT, A. C. 1978. Sedimentological and ecological 2006] DIMICHELE ET AL.: PTERIDOSPERM PALEOECOLOGY 117

control of Westphalian B plant assemblages from STEWART, W N. AND T DELEVORYAS. 1956. The med- west Yorkshire. Proc. Yorkshire Geol. Soc. 41: uUosan pteridosperms. Bot. Rev. 22: 45—80. 461-508. STIDD, B. M. 1990. Further documentation of the struc- SCOTT, A. C. 1984. Studies on the sedimentology, pa- ture of Dolerotheca and a critique of other theories. laeontology and palaeoecology of the Middle Coal Palaeontographica 217B: 51—86. Measures (Wesphalian B, Upper Carboniferous) at STIDD, B. M. AND T L. PHILLIPS. 1973. The vegetative Swillington, Yorkshire. I. Introduction. Trans. anatomy of Schopfiastrum decussatum from the Leeds Geol. Assoc. 10: 1—16. Middle Pennsylvanian of the Illinois Basin. Am. J. SCOTT, A. C. AND J. GALTIER. 1985. Distribution and Bot. 60: 463-474. ecology of early ferns. Proc. Roy. Soc. Edinburgh STIDD, B. M. AND T L. PHILLIPS. 1982. Johnhallia la- 86 B: 141-149. cunosa gen. et sp. n.: a new pteridosperm from the SCOTT, A. C. AND J. GALTIER. 1988. A new Lower Middle Pennylvanian of Indiana. J. Paleontol. 56: Carboniferous flora from East Lothian, Scotland. 1093-1102. Proceedings of the Geologists' Association 99: STIDD, B. M., L. L. OESTRY, AND T L. PHILLIPS. 1975. 141-151. On the frond of Sutcliffia insignis var. tuberculata. SCOTT, A. C. AND B. MEYER-BERTHAUD. 1985. Plants Rev. Palaeobot. Palynol. 20: 55-66. from the Dinatian of Foulden, Berwickshire, Scot- SUN B., G. SHEN, AND X. HE. 1999. New material of land. Trans. Roy. Soc. Edinburgh, Earth Sciences Protoblechnum from Permian coal-bearing strata in 76: 13-20. western Henan. Earth Science Journal of China SCOTT, A. C, J. GALTIER, AND G. CLAYTON. 1984. Dis- University of Geosciences 24: 161—166 (in Chinese tribution of anatomically-preserved floras in the with English summary). Lower Carboniferous in Western Europe. Trans. TABOR, N. J. AND I. P. MONTANEZ. 2004. Morphology Roy. Soc. Edinburgh, Earth Sciences 75: 311—340. and distribution of fossil soils in the Permo-Penn- SCOTT, A. C, R. BROWN, J. GALTIER, AND B. MEYER- sylvanian Bowie and Wichita Groups, North-Cen- BERTHAUD. 1994. Fossil plants from the Visean of tral Texas, USA: their implications for western East Kirkton, West Lothian, Scotland. Trans. Roy. equatorial Pangean palaeoclimate during icehouse- Soc. Edinburgh, Earth Sciences 84: 249—260. greenhouse transition. Sedimentology 51: 851—884. TABOR, N. AND MONTANEZ. SCOTT, D. H. 1899. On the structure and affinities of J. I. P. 2005. Oxygen and fossil plants from the Palaeozoic rocks. IIL Med- hydrogen isotope compositions of Permian pedo- ullosa anglica, a new representative of the Cyca- genic phyllosilicates: development of modern sur- dofilices. Phil. Trans. Roy. Soc. London, B 191: face domain arrays and implications for paletem- perature reconstructions. Palaeogeography, Palaeo- 81-126. climatology, Palaeoecology 223: 127—146. SELLARDS, E. H. 1900. A new genus of ferns from the TAYLOR, T N. 1965. Paleozoic seed studies: a mono- Permian of Kansas. Kansas University Quarterly 9: graph of the American species of Pachytesta. Pa- 179-189. laeontographica Abt. B: 117: 1—46. SERBET, R. AND G. W ROTHWELL. 1992. Characterizing TAYLOR, T N. 1978. The ultrastructure and reproduc- the most primitive seed ferns. I. a reconstruction of tive significance of Monoletes (Pteridospermales) Elkinsia polymorpha. Int. J. Plant Sci. 153: 602— pollen. Can. J. Bot 56: 3105-3118. 621. TAYLOR, T. N. 1982. Reproductive biology in early SERBET, R. AND G. W ROTHWELL. 1995. Functional seed plants. Bioscience 32: 23—28. morphology and homologies of gymnospermous TAYLOR, T. N. AND D. A. EGGERT. 1967. Petrified ovules: evidence from a new species of Stephan- plants from the Upper Mississippian (Chester Se- ospermum (MeduUosales). Can. J. Bot 73: 650— ries) of Arkansas. Trans. Am. Microscopical Soc. 661. 86: 412-416. SHADLE, G. L. AND B. M. STIDD. 1975. The frond of TAYLOR, T. N. AND M. A. MILLAY. 1981. Additional Heterangium. Am. J. Bot. 62: 67—75. information on the pollen organ Halletheca (Med- SHUTE, C. H. AND C. J. CLEAL. 2002. Ecology and uUosales). Am. J. Bot. 68: 1403-1407. growth habit of Laveineopteris: a gymnosperm ToMEScu, A. M. F, G. W. ROTHWELL, AND G. MAPES. from the Late Carboniferous tropical rain forests. 2001. Lyginopteris royalii sp. nov. from the Upper Palaeontology 45: 943-972. Mississippian of North America. Rev. Palaeobot. SiMUNEK, Z. AND C. J. CLEAL. 2004. Small-pinnuled Palynol. 116: 159-173. odontopterid meduUosaleans from the middle and VALENTINE, J. W. 1980. Determinants of diversity in upper Stephanian of Bohemia and Saar-Lorraine. higher taxonomic categories. Paleobiology 6: 444— Rev. Palaeobot. Palynol. 129: 21-38. 450. SPECK, T 1994. A biomechanical method to distinguish WANG Z.-Q. 1997. Permian Supaia fronds and an as- between self-supporting and non-self-supporting sociated Autunia fructification from Shanxi, China. fossil plants. Rev. Palaeobot. Palynol. 81: 65—82. Palaeontology 40: 245-277. SPECK, T AND N. P. ROWE. 1994. Biomechanical anal- WHITE, D. 1929. Flora of the Hermit Shale, Grand ysis of Pitus dayi: early seed plant vegetative mor- Canyon, Arizona. Carnegie Institution of Washing- phology and its implications on growth habit. J. ton, Publication No. 405. 221 p. Plant Res. 107: 443-460. WHITE, D. 1943. Lower Pennsylvanian species of Mar- STERZEL, J. T 1918. Die organischen Reste des Kulms iopteris, Eremoptejis, Diplothmema, and Aneimites und Rotliegenden der Gegend von Chemnitz. Kon- from the Appalachian region U.S. Geological Sur- iglich Sachsischen Gesellschaft der Wissenschaf- vey Professional Paper 197-C: 85-140. ten, Abhandlungen Mathmatisch-Physischen Klas- WiLLARD, D. A., W A. DiMlCHELE, J. C. HOWER, D. se 35: 205-315. L. EGGERT, C. B. REXROAD, AND A. C. SCOTT. 1995. 118 JOURNAL OF THE TORREY BOTANICAL SOCIETY [VOL. 133

Paleoecology of the Springfield Coal Member ZEILLER, R. 1888. Etudes sur le terrain houiller de (Middle Pennsylvanian, Illinois Basin) near the Commentry, 366 pp. In R. Schropp [ed.], Flore Leslie Cemetery Paleochannel, southwestern Indi- Fossile, Bulletin de la Societe de I'industrie Mi- ana. Int. J. Coal Geol. 27: 59-98. nerale, 3"". Ser 2: 1-366. WINSTON, R. B. AND T. L. PHILLIPS. 1991. A structur- ZlEGLER, A. M. 1990. Phytogeographic patterns and ally preserved. Lower Pennsylvanian flora from the continental configurations during the Permian Pe- New Castle coal bed of Alabama. Geol. Survey of riod, p. 363-379. In C. McKerrow and C. R. Sco- Alabama Circular 157: 1—21. tese [Eds.], Palaeozoic Palaeogeography and Bio- WNUK, C. AND H. W. PFEFFERKORN. 1984. The life hab- geography. Geol. Soc. London Memoir 12. its and paleoecology of Middle Pennsylvanian pte- ZODROW, E. L. AND C. J. CLEAL. 1993. The epidermal ridosperms based on an in situ assemblage from the Bernice Basin (Sullivan County, Pennsylvania, structure of the Carboniferous gymnosperm frond U.S.A.). Rev. Palaeobot. Palynol. 41: 329-351. Reticulopteiis Gothan. Palaeontology 36: 65—97. WNUK, C. AND H. W. PFEFFERKORN. 1987. A Pennsyl- ZoDROW, E. L. AND C. J. CLEAL. 1998. Revision of the vanian-age terrestrial storm deposit: using plant pteridosperm foliage Alethopteris and Lonchopter- fossils to characterize the history and process of idium (Upper Carboniferous), Sydney Coalfield, sediment accumulation. J. Sedimentary Petrology Nova Scotia, Canada. Palaeontographica B 247: 57: 212-221. 65-122.